rfeedparser 0.9.87 → 0.9.91
Sign up to get free protection for your applications and to get access to all the features.
- data/lib/rfeedparser.rb +37 -3191
- data/lib/rfeedparser/aliases.rb +432 -0
- data/lib/rfeedparser/better_attributelist.rb +41 -0
- data/lib/rfeedparser/better_sgmlparser.rb +264 -0
- data/lib/rfeedparser/encoding_helpers.rb +257 -0
- data/lib/rfeedparser/feedparserdict.rb +93 -0
- data/lib/rfeedparser/forgiving_uri.rb +947 -0
- data/lib/rfeedparser/markup_helpers.rb +74 -0
- data/lib/rfeedparser/parser_mixin.rb +1235 -0
- data/lib/rfeedparser/parsers.rb +177 -0
- data/lib/rfeedparser/scrub.rb +207 -0
- data/lib/rfeedparser/time_helpers.rb +408 -0
- data/tests/rfeedparsertest.rb +3 -1
- metadata +3270 -3249
data/lib/rfeedparser.rb
CHANGED
|
@@ -14,22 +14,21 @@ require 'stringio'
|
|
|
14
14
|
require 'uri'
|
|
15
15
|
require 'cgi' # escaping html
|
|
16
16
|
require 'time'
|
|
17
|
-
require 'xml/saxdriver' # calling expat
|
|
18
17
|
require 'pp'
|
|
19
18
|
require 'rubygems'
|
|
20
19
|
require 'base64'
|
|
21
20
|
require 'iconv'
|
|
22
|
-
|
|
21
|
+
|
|
23
22
|
gem 'character-encodings', ">=0.2.0"
|
|
24
23
|
gem 'htmltools', ">=1.10"
|
|
25
24
|
gem 'htmlentities', ">=4.0.0"
|
|
26
25
|
gem 'activesupport', ">=1.4.1"
|
|
27
26
|
gem 'rchardet', ">=1.0"
|
|
27
|
+
require 'xml/saxdriver' # calling expat through the xmlparser gem
|
|
28
28
|
|
|
29
29
|
require 'rchardet'
|
|
30
30
|
$chardet = true
|
|
31
31
|
|
|
32
|
-
require 'hpricot'
|
|
33
32
|
require 'encoding/character/utf-8'
|
|
34
33
|
require 'html/sgml-parser'
|
|
35
34
|
require 'htmlentities'
|
|
@@ -40,998 +39,24 @@ include OpenURI
|
|
|
40
39
|
$debug = false
|
|
41
40
|
$compatible = true
|
|
42
41
|
|
|
43
|
-
|
|
44
|
-
|
|
45
|
-
|
|
46
|
-
|
|
47
|
-
|
|
48
|
-
|
|
49
|
-
|
|
50
|
-
|
|
51
|
-
|
|
52
|
-
|
|
53
|
-
|
|
54
|
-
|
|
55
|
-
'646' => 'ascii',
|
|
56
|
-
'ansi_x3.4_1968' => 'ascii',
|
|
57
|
-
'ansi_x3_4_1968' => 'ascii', # some email headers use this non-standard name
|
|
58
|
-
'ansi_x3.4_1986' => 'ascii',
|
|
59
|
-
'cp367' => 'ascii',
|
|
60
|
-
'csascii' => 'ascii',
|
|
61
|
-
'ibm367' => 'ascii',
|
|
62
|
-
'iso646_us' => 'ascii',
|
|
63
|
-
'iso_646.irv_1991' => 'ascii',
|
|
64
|
-
'iso_ir_6' => 'ascii',
|
|
65
|
-
'us' => 'ascii',
|
|
66
|
-
'us_ascii' => 'ascii',
|
|
67
|
-
|
|
68
|
-
# big5 codec
|
|
69
|
-
'big5_tw' => 'big5',
|
|
70
|
-
'csbig5' => 'big5',
|
|
71
|
-
|
|
72
|
-
# big5hkscs codec
|
|
73
|
-
'big5_hkscs' => 'big5hkscs',
|
|
74
|
-
'hkscs' => 'big5hkscs',
|
|
75
|
-
|
|
76
|
-
# cp037 codec
|
|
77
|
-
'037' => 'cp037',
|
|
78
|
-
'csibm037' => 'cp037',
|
|
79
|
-
'ebcdic_cp_ca' => 'cp037',
|
|
80
|
-
'ebcdic_cp_nl' => 'cp037',
|
|
81
|
-
'ebcdic_cp_us' => 'cp037',
|
|
82
|
-
'ebcdic_cp_wt' => 'cp037',
|
|
83
|
-
'ibm037' => 'cp037',
|
|
84
|
-
'ibm039' => 'cp037',
|
|
85
|
-
|
|
86
|
-
# cp1026 codec
|
|
87
|
-
'1026' => 'cp1026',
|
|
88
|
-
'csibm1026' => 'cp1026',
|
|
89
|
-
'ibm1026' => 'cp1026',
|
|
90
|
-
|
|
91
|
-
# cp1140 codec
|
|
92
|
-
'1140' => 'cp1140',
|
|
93
|
-
'ibm1140' => 'cp1140',
|
|
94
|
-
|
|
95
|
-
# cp1250 codec
|
|
96
|
-
'1250' => 'cp1250',
|
|
97
|
-
'windows_1250' => 'cp1250',
|
|
98
|
-
|
|
99
|
-
# cp1251 codec
|
|
100
|
-
'1251' => 'cp1251',
|
|
101
|
-
'windows_1251' => 'cp1251',
|
|
102
|
-
|
|
103
|
-
# cp1252 codec
|
|
104
|
-
'1252' => 'cp1252',
|
|
105
|
-
'windows_1252' => 'cp1252',
|
|
106
|
-
|
|
107
|
-
# cp1253 codec
|
|
108
|
-
'1253' => 'cp1253',
|
|
109
|
-
'windows_1253' => 'cp1253',
|
|
110
|
-
|
|
111
|
-
# cp1254 codec
|
|
112
|
-
'1254' => 'cp1254',
|
|
113
|
-
'windows_1254' => 'cp1254',
|
|
114
|
-
|
|
115
|
-
# cp1255 codec
|
|
116
|
-
'1255' => 'cp1255',
|
|
117
|
-
'windows_1255' => 'cp1255',
|
|
118
|
-
|
|
119
|
-
# cp1256 codec
|
|
120
|
-
'1256' => 'cp1256',
|
|
121
|
-
'windows_1256' => 'cp1256',
|
|
122
|
-
|
|
123
|
-
# cp1257 codec
|
|
124
|
-
'1257' => 'cp1257',
|
|
125
|
-
'windows_1257' => 'cp1257',
|
|
126
|
-
|
|
127
|
-
# cp1258 codec
|
|
128
|
-
'1258' => 'cp1258',
|
|
129
|
-
'windows_1258' => 'cp1258',
|
|
130
|
-
|
|
131
|
-
# cp424 codec
|
|
132
|
-
'424' => 'cp424',
|
|
133
|
-
'csibm424' => 'cp424',
|
|
134
|
-
'ebcdic_cp_he' => 'cp424',
|
|
135
|
-
'ibm424' => 'cp424',
|
|
136
|
-
|
|
137
|
-
# cp437 codec
|
|
138
|
-
'437' => 'cp437',
|
|
139
|
-
'cspc8codepage437' => 'cp437',
|
|
140
|
-
'ibm437' => 'cp437',
|
|
141
|
-
|
|
142
|
-
# cp500 codec
|
|
143
|
-
'500' => 'cp500',
|
|
144
|
-
'csibm500' => 'cp500',
|
|
145
|
-
'ebcdic_cp_be' => 'cp500',
|
|
146
|
-
'ebcdic_cp_ch' => 'cp500',
|
|
147
|
-
'ibm500' => 'cp500',
|
|
148
|
-
|
|
149
|
-
# cp775 codec
|
|
150
|
-
'775' => 'cp775',
|
|
151
|
-
'cspc775baltic' => 'cp775',
|
|
152
|
-
'ibm775' => 'cp775',
|
|
153
|
-
|
|
154
|
-
# cp850 codec
|
|
155
|
-
'850' => 'cp850',
|
|
156
|
-
'cspc850multilingual' => 'cp850',
|
|
157
|
-
'ibm850' => 'cp850',
|
|
158
|
-
|
|
159
|
-
# cp852 codec
|
|
160
|
-
'852' => 'cp852',
|
|
161
|
-
'cspcp852' => 'cp852',
|
|
162
|
-
'ibm852' => 'cp852',
|
|
163
|
-
|
|
164
|
-
# cp855 codec
|
|
165
|
-
'855' => 'cp855',
|
|
166
|
-
'csibm855' => 'cp855',
|
|
167
|
-
'ibm855' => 'cp855',
|
|
168
|
-
|
|
169
|
-
# cp857 codec
|
|
170
|
-
'857' => 'cp857',
|
|
171
|
-
'csibm857' => 'cp857',
|
|
172
|
-
'ibm857' => 'cp857',
|
|
173
|
-
|
|
174
|
-
# cp860 codec
|
|
175
|
-
'860' => 'cp860',
|
|
176
|
-
'csibm860' => 'cp860',
|
|
177
|
-
'ibm860' => 'cp860',
|
|
178
|
-
|
|
179
|
-
# cp861 codec
|
|
180
|
-
'861' => 'cp861',
|
|
181
|
-
'cp_is' => 'cp861',
|
|
182
|
-
'csibm861' => 'cp861',
|
|
183
|
-
'ibm861' => 'cp861',
|
|
184
|
-
|
|
185
|
-
# cp862 codec
|
|
186
|
-
'862' => 'cp862',
|
|
187
|
-
'cspc862latinhebrew' => 'cp862',
|
|
188
|
-
'ibm862' => 'cp862',
|
|
189
|
-
|
|
190
|
-
# cp863 codec
|
|
191
|
-
'863' => 'cp863',
|
|
192
|
-
'csibm863' => 'cp863',
|
|
193
|
-
'ibm863' => 'cp863',
|
|
194
|
-
|
|
195
|
-
# cp864 codec
|
|
196
|
-
'864' => 'cp864',
|
|
197
|
-
'csibm864' => 'cp864',
|
|
198
|
-
'ibm864' => 'cp864',
|
|
199
|
-
|
|
200
|
-
# cp865 codec
|
|
201
|
-
'865' => 'cp865',
|
|
202
|
-
'csibm865' => 'cp865',
|
|
203
|
-
'ibm865' => 'cp865',
|
|
204
|
-
|
|
205
|
-
# cp866 codec
|
|
206
|
-
'866' => 'cp866',
|
|
207
|
-
'csibm866' => 'cp866',
|
|
208
|
-
'ibm866' => 'cp866',
|
|
209
|
-
|
|
210
|
-
# cp869 codec
|
|
211
|
-
'869' => 'cp869',
|
|
212
|
-
'cp_gr' => 'cp869',
|
|
213
|
-
'csibm869' => 'cp869',
|
|
214
|
-
'ibm869' => 'cp869',
|
|
215
|
-
|
|
216
|
-
# cp932 codec
|
|
217
|
-
'932' => 'cp932',
|
|
218
|
-
'ms932' => 'cp932',
|
|
219
|
-
'mskanji' => 'cp932',
|
|
220
|
-
'ms_kanji' => 'cp932',
|
|
221
|
-
|
|
222
|
-
# cp949 codec
|
|
223
|
-
'949' => 'cp949',
|
|
224
|
-
'ms949' => 'cp949',
|
|
225
|
-
'uhc' => 'cp949',
|
|
226
|
-
|
|
227
|
-
# cp950 codec
|
|
228
|
-
'950' => 'cp950',
|
|
229
|
-
'ms950' => 'cp950',
|
|
230
|
-
|
|
231
|
-
# euc_jp codec
|
|
232
|
-
'euc_jp' => 'euc-jp',
|
|
233
|
-
'eucjp' => 'euc-jp',
|
|
234
|
-
'ujis' => 'euc-jp',
|
|
235
|
-
'u_jis' => 'euc-jp',
|
|
236
|
-
|
|
237
|
-
# euc_kr codec
|
|
238
|
-
'euc_kr' => 'euc-kr',
|
|
239
|
-
'euckr' => 'euc-kr',
|
|
240
|
-
'korean' => 'euc-kr',
|
|
241
|
-
'ksc5601' => 'euc-kr',
|
|
242
|
-
'ks_c_5601' => 'euc-kr',
|
|
243
|
-
'ks_c_5601_1987' => 'euc-kr',
|
|
244
|
-
'ksx1001' => 'euc-kr',
|
|
245
|
-
'ks_x_1001' => 'euc-kr',
|
|
246
|
-
|
|
247
|
-
# gb18030 codec
|
|
248
|
-
'gb18030_2000' => 'gb18030',
|
|
249
|
-
|
|
250
|
-
# gb2312 codec
|
|
251
|
-
'chinese' => 'gb2312',
|
|
252
|
-
'csiso58gb231280' => 'gb2312',
|
|
253
|
-
'euc_cn' => 'gb2312',
|
|
254
|
-
'euccn' => 'gb2312',
|
|
255
|
-
'eucgb2312_cn' => 'gb2312',
|
|
256
|
-
'gb2312_1980' => 'gb2312',
|
|
257
|
-
'gb2312_80' => 'gb2312',
|
|
258
|
-
'iso_ir_58' => 'gb2312',
|
|
259
|
-
|
|
260
|
-
# gbk codec
|
|
261
|
-
'936' => 'gbk',
|
|
262
|
-
'cp936' => 'gbk',
|
|
263
|
-
'ms936' => 'gbk',
|
|
264
|
-
|
|
265
|
-
# hp-roman8 codec
|
|
266
|
-
'hp_roman8' => 'hp-roman8',
|
|
267
|
-
'roman8' => 'hp-roman8',
|
|
268
|
-
'r8' => 'hp-roman8',
|
|
269
|
-
'csHPRoman8' => 'hp-roman8',
|
|
270
|
-
|
|
271
|
-
# iso2022_jp codec
|
|
272
|
-
'iso2022_jp' => 'iso-2022-jp',
|
|
273
|
-
'csiso2022jp' => 'iso-2022-jp',
|
|
274
|
-
'iso2022jp' => 'iso-2022-jp',
|
|
275
|
-
'iso_2022_jp' => 'iso-2022-jp',
|
|
276
|
-
|
|
277
|
-
# iso2022_jp_1 codec
|
|
278
|
-
'iso2002_jp_1' => 'iso-2022-jp-1',
|
|
279
|
-
'iso2022jp_1' => 'iso-2022-jp-1',
|
|
280
|
-
'iso_2022_jp_1' => 'iso-2022-jp-1',
|
|
281
|
-
|
|
282
|
-
# iso2022_jp_2 codec
|
|
283
|
-
'iso2022_jp_2' => 'iso-2002-jp-2',
|
|
284
|
-
'iso2022jp_2' => 'iso-2022-jp-2',
|
|
285
|
-
'iso_2022_jp_2' => 'iso-2022-jp-2',
|
|
286
|
-
|
|
287
|
-
# iso2022_jp_3 codec
|
|
288
|
-
'iso2002_jp_3' => 'iso-2022-jp-3',
|
|
289
|
-
'iso2022jp_3' => 'iso-2022-jp-3',
|
|
290
|
-
'iso_2022_jp_3' => 'iso-2022-jp-3',
|
|
291
|
-
|
|
292
|
-
# iso2022_kr codec
|
|
293
|
-
'iso2022_kr' => 'iso-2022-kr',
|
|
294
|
-
'csiso2022kr' => 'iso-2022-kr',
|
|
295
|
-
'iso2022kr' => 'iso-2022-kr',
|
|
296
|
-
'iso_2022_kr' => 'iso-2022-kr',
|
|
297
|
-
|
|
298
|
-
# iso8859_10 codec
|
|
299
|
-
'iso8859_10' => 'iso-8859-10',
|
|
300
|
-
'csisolatin6' => 'iso-8859-10',
|
|
301
|
-
'iso_8859_10' => 'iso-8859-10',
|
|
302
|
-
'iso_8859_10_1992' => 'iso-8859-10',
|
|
303
|
-
'iso_ir_157' => 'iso-8859-10',
|
|
304
|
-
'l6' => 'iso-8859-10',
|
|
305
|
-
'latin6' => 'iso-8859-10',
|
|
306
|
-
|
|
307
|
-
# iso8859_13 codec
|
|
308
|
-
'iso8859_13' => 'iso-8859-13',
|
|
309
|
-
'iso_8859_13' => 'iso-8859-13',
|
|
310
|
-
|
|
311
|
-
# iso8859_14 codec
|
|
312
|
-
'iso8859_14' => 'iso-8859-14',
|
|
313
|
-
'iso_8859_14' => 'iso-8859-14',
|
|
314
|
-
'iso_8859_14_1998' => 'iso-8859-14',
|
|
315
|
-
'iso_celtic' => 'iso-8859-14',
|
|
316
|
-
'iso_ir_199' => 'iso-8859-14',
|
|
317
|
-
'l8' => 'iso-8859-14',
|
|
318
|
-
'latin8' => 'iso-8859-14',
|
|
319
|
-
|
|
320
|
-
# iso8859_15 codec
|
|
321
|
-
'iso8859_15' => 'iso-8859-15',
|
|
322
|
-
'iso_8859_15' => 'iso-8859-15',
|
|
323
|
-
|
|
324
|
-
# iso8859_1 codec
|
|
325
|
-
'latin_1' => 'iso-8859-1',
|
|
326
|
-
'cp819' => 'iso-8859-1',
|
|
327
|
-
'csisolatin1' => 'iso-8859-1',
|
|
328
|
-
'ibm819' => 'iso-8859-1',
|
|
329
|
-
'iso8859' => 'iso-8859-1',
|
|
330
|
-
'iso_8859_1' => 'iso-8859-1',
|
|
331
|
-
'iso_8859_1_1987' => 'iso-8859-1',
|
|
332
|
-
'iso_ir_100' => 'iso-8859-1',
|
|
333
|
-
'l1' => 'iso-8859-1',
|
|
334
|
-
'latin' => 'iso-8859-1',
|
|
335
|
-
'latin1' => 'iso-8859-1',
|
|
336
|
-
|
|
337
|
-
# iso8859_2 codec
|
|
338
|
-
'iso8859_2' => 'iso-8859-2',
|
|
339
|
-
'csisolatin2' => 'iso-8859-2',
|
|
340
|
-
'iso_8859_2' => 'iso-8859-2',
|
|
341
|
-
'iso_8859_2_1987' => 'iso-8859-2',
|
|
342
|
-
'iso_ir_101' => 'iso-8859-2',
|
|
343
|
-
'l2' => 'iso-8859-2',
|
|
344
|
-
'latin2' => 'iso-8859-2',
|
|
345
|
-
|
|
346
|
-
# iso8859_3 codec
|
|
347
|
-
'iso8859_3' => 'iso-8859-3',
|
|
348
|
-
'csisolatin3' => 'iso-8859-3',
|
|
349
|
-
'iso_8859_3' => 'iso-8859-3',
|
|
350
|
-
'iso_8859_3_1988' => 'iso-8859-3',
|
|
351
|
-
'iso_ir_109' => 'iso-8859-3',
|
|
352
|
-
'l3' => 'iso-8859-3',
|
|
353
|
-
'latin3' => 'iso-8859-3',
|
|
354
|
-
|
|
355
|
-
# iso8859_4 codec
|
|
356
|
-
'iso8849_4' => 'iso-8859-4',
|
|
357
|
-
'csisolatin4' => 'iso-8859-4',
|
|
358
|
-
'iso_8859_4' => 'iso-8859-4',
|
|
359
|
-
'iso_8859_4_1988' => 'iso-8859-4',
|
|
360
|
-
'iso_ir_110' => 'iso-8859-4',
|
|
361
|
-
'l4' => 'iso-8859-4',
|
|
362
|
-
'latin4' => 'iso-8859-4',
|
|
363
|
-
|
|
364
|
-
# iso8859_5 codec
|
|
365
|
-
'iso8859_5' => 'iso-8859-5',
|
|
366
|
-
'csisolatincyrillic' => 'iso-8859-5',
|
|
367
|
-
'cyrillic' => 'iso-8859-5',
|
|
368
|
-
'iso_8859_5' => 'iso-8859-5',
|
|
369
|
-
'iso_8859_5_1988' => 'iso-8859-5',
|
|
370
|
-
'iso_ir_144' => 'iso-8859-5',
|
|
371
|
-
|
|
372
|
-
# iso8859_6 codec
|
|
373
|
-
'iso8859_6' => 'iso-8859-6',
|
|
374
|
-
'arabic' => 'iso-8859-6',
|
|
375
|
-
'asmo_708' => 'iso-8859-6',
|
|
376
|
-
'csisolatinarabic' => 'iso-8859-6',
|
|
377
|
-
'ecma_114' => 'iso-8859-6',
|
|
378
|
-
'iso_8859_6' => 'iso-8859-6',
|
|
379
|
-
'iso_8859_6_1987' => 'iso-8859-6',
|
|
380
|
-
'iso_ir_127' => 'iso-8859-6',
|
|
381
|
-
|
|
382
|
-
# iso8859_7 codec
|
|
383
|
-
'iso8859_7' => 'iso-8859-7',
|
|
384
|
-
'csisolatingreek' => 'iso-8859-7',
|
|
385
|
-
'ecma_118' => 'iso-8859-7',
|
|
386
|
-
'elot_928' => 'iso-8859-7',
|
|
387
|
-
'greek' => 'iso-8859-7',
|
|
388
|
-
'greek8' => 'iso-8859-7',
|
|
389
|
-
'iso_8859_7' => 'iso-8859-7',
|
|
390
|
-
'iso_8859_7_1987' => 'iso-8859-7',
|
|
391
|
-
'iso_ir_126' => 'iso-8859-7',
|
|
392
|
-
|
|
393
|
-
# iso8859_8 codec
|
|
394
|
-
'iso8859_9' => 'iso8859_8',
|
|
395
|
-
'csisolatinhebrew' => 'iso-8859-8',
|
|
396
|
-
'hebrew' => 'iso-8859-8',
|
|
397
|
-
'iso_8859_8' => 'iso-8859-8',
|
|
398
|
-
'iso_8859_8_1988' => 'iso-8859-8',
|
|
399
|
-
'iso_ir_138' => 'iso-8859-8',
|
|
400
|
-
|
|
401
|
-
# iso8859_9 codec
|
|
402
|
-
'iso8859_9' => 'iso-8859-9',
|
|
403
|
-
'csisolatin5' => 'iso-8859-9',
|
|
404
|
-
'iso_8859_9' => 'iso-8859-9',
|
|
405
|
-
'iso_8859_9_1989' => 'iso-8859-9',
|
|
406
|
-
'iso_ir_148' => 'iso-8859-9',
|
|
407
|
-
'l5' => 'iso-8859-9',
|
|
408
|
-
'latin5' => 'iso-8859-9',
|
|
409
|
-
|
|
410
|
-
# iso8859_11 codec
|
|
411
|
-
'iso8859_11' => 'iso-8859-11',
|
|
412
|
-
'thai' => 'iso-8859-11',
|
|
413
|
-
'iso_8859_11' => 'iso-8859-11',
|
|
414
|
-
'iso_8859_11_2001' => 'iso-8859-11',
|
|
415
|
-
|
|
416
|
-
# iso8859_16 codec
|
|
417
|
-
'iso8859_16' => 'iso-8859-16',
|
|
418
|
-
'iso_8859_16' => 'iso-8859-16',
|
|
419
|
-
'iso_8859_16_2001' => 'iso-8859-16',
|
|
420
|
-
'iso_ir_226' => 'iso-8859-16',
|
|
421
|
-
'l10' => 'iso-8859-16',
|
|
422
|
-
'latin10' => 'iso-8859-16',
|
|
423
|
-
|
|
424
|
-
# cskoi8r codec
|
|
425
|
-
'koi8_r' => 'cskoi8r',
|
|
426
|
-
|
|
427
|
-
# mac_cyrillic codec
|
|
428
|
-
'mac_cyrillic' => 'maccyrillic',
|
|
429
|
-
|
|
430
|
-
# shift_jis codec
|
|
431
|
-
'csshiftjis' => 'shift_jis',
|
|
432
|
-
'shiftjis' => 'shift_jis',
|
|
433
|
-
'sjis' => 'shift_jis',
|
|
434
|
-
's_jis' => 'shift_jis',
|
|
435
|
-
|
|
436
|
-
# shift_jisx0213 codec
|
|
437
|
-
'shiftjisx0213' => 'shift_jisx0213',
|
|
438
|
-
'sjisx0213' => 'shift_jisx0213',
|
|
439
|
-
's_jisx0213' => 'shift_jisx0213',
|
|
440
|
-
|
|
441
|
-
# utf_16 codec
|
|
442
|
-
'utf_16' => 'utf-16',
|
|
443
|
-
'u16' => 'utf-16',
|
|
444
|
-
'utf16' => 'utf-16',
|
|
445
|
-
|
|
446
|
-
# utf_16_be codec
|
|
447
|
-
'utf_16_be' => 'utf-16be',
|
|
448
|
-
'unicodebigunmarked' => 'utf-16be',
|
|
449
|
-
'utf_16be' => 'utf-16be',
|
|
450
|
-
|
|
451
|
-
# utf_16_le codec
|
|
452
|
-
'utf_16_le' => 'utf-16le',
|
|
453
|
-
'unicodelittleunmarked' => 'utf-16le',
|
|
454
|
-
'utf_16le' => 'utf-16le',
|
|
455
|
-
|
|
456
|
-
# utf_7 codec
|
|
457
|
-
'utf_7' => 'utf-7',
|
|
458
|
-
'u7' => 'utf-7',
|
|
459
|
-
'utf7' => 'utf-7',
|
|
460
|
-
|
|
461
|
-
# utf_8 codec
|
|
462
|
-
'utf_8' => 'utf-8',
|
|
463
|
-
'u8' => 'utf-8',
|
|
464
|
-
'utf' => 'utf-8',
|
|
465
|
-
'utf8' => 'utf-8',
|
|
466
|
-
'utf8_ucs2' => 'utf-8',
|
|
467
|
-
'utf8_ucs4' => 'utf-8',
|
|
468
|
-
}
|
|
469
|
-
|
|
470
|
-
def unicode(data, from_encoding)
|
|
471
|
-
# Takes a single string and converts it from the encoding in
|
|
472
|
-
# from_encoding to unicode.
|
|
473
|
-
uconvert(data, from_encoding, 'unicode')
|
|
474
|
-
end
|
|
475
|
-
|
|
476
|
-
def uconvert(data, from_encoding, to_encoding = 'utf-8')
|
|
477
|
-
from_encoding = Encoding_Aliases[from_encoding] || from_encoding
|
|
478
|
-
to_encoding = Encoding_Aliases[to_encoding] || to_encoding
|
|
479
|
-
Iconv.iconv(to_encoding, from_encoding, data)[0]
|
|
480
|
-
end
|
|
481
|
-
|
|
482
|
-
def unichr(i)
|
|
483
|
-
[i].pack('U*')
|
|
484
|
-
end
|
|
485
|
-
|
|
486
|
-
def index_match(stri,regexp, offset)
|
|
487
|
-
i = stri.index(regexp, offset)
|
|
488
|
-
|
|
489
|
-
return nil, nil unless i
|
|
490
|
-
|
|
491
|
-
full = stri[i..-1].match(regexp)
|
|
492
|
-
return i, full
|
|
493
|
-
end
|
|
494
|
-
|
|
495
|
-
def _ebcdic_to_ascii(s)
|
|
496
|
-
return Iconv.iconv("iso88591", "ebcdic-cp-be", s)[0]
|
|
497
|
-
end
|
|
498
|
-
|
|
499
|
-
def urljoin(base, uri)
|
|
500
|
-
urifixer = /^([A-Za-z][A-Za-z0-9+-.]*:\/\/)(\/*)(.*?)/u
|
|
501
|
-
uri = uri.sub(urifixer, '\1\3')
|
|
502
|
-
|
|
503
|
-
begin
|
|
504
|
-
return URI.join(base, uri).to_s
|
|
505
|
-
rescue URI::BadURIError => e
|
|
506
|
-
if URI.parse(base).relative?
|
|
507
|
-
return URI::parse(uri).to_s
|
|
508
|
-
end
|
|
509
|
-
end
|
|
510
|
-
end
|
|
511
|
-
|
|
512
|
-
def py2rtime(pytuple)
|
|
513
|
-
Time.utc(pytuple[0..5])
|
|
514
|
-
end
|
|
515
|
-
|
|
516
|
-
# http://intertwingly.net/stories/2005/09/28/xchar.rb
|
|
517
|
-
module XChar
|
|
518
|
-
# http://intertwingly.net/stories/2004/04/14/i18n.html#CleaningWindows
|
|
519
|
-
CP1252 = {
|
|
520
|
-
128 => 8364, # euro sign
|
|
521
|
-
130 => 8218, # single low-9 quotation mark
|
|
522
|
-
131 => 402, # latin small letter f with hook
|
|
523
|
-
132 => 8222, # double low-9 quotation mark
|
|
524
|
-
133 => 8230, # horizontal ellipsis
|
|
525
|
-
134 => 8224, # dagger
|
|
526
|
-
135 => 8225, # double dagger
|
|
527
|
-
136 => 710, # modifier letter circumflex accent
|
|
528
|
-
137 => 8240, # per mille sign
|
|
529
|
-
138 => 352, # latin capital letter s with caron
|
|
530
|
-
139 => 8249, # single left-pointing angle quotation mark
|
|
531
|
-
140 => 338, # latin capital ligature oe
|
|
532
|
-
142 => 381, # latin capital letter z with caron
|
|
533
|
-
145 => 8216, # left single quotation mark
|
|
534
|
-
146 => 8217, # right single quotation mark
|
|
535
|
-
147 => 8220, # left double quotation mark
|
|
536
|
-
148 => 8221, # right double quotation mark
|
|
537
|
-
149 => 8226, # bullet
|
|
538
|
-
150 => 8211, # en dash
|
|
539
|
-
151 => 8212, # em dash
|
|
540
|
-
152 => 732, # small tilde
|
|
541
|
-
153 => 8482, # trade mark sign
|
|
542
|
-
154 => 353, # latin small letter s with caron
|
|
543
|
-
155 => 8250, # single right-pointing angle quotation mark
|
|
544
|
-
156 => 339, # latin small ligature oe
|
|
545
|
-
158 => 382, # latin small letter z with caron
|
|
546
|
-
159 => 376} # latin capital letter y with diaeresis
|
|
547
|
-
|
|
548
|
-
# http://www.w3.org/TR/REC-xml/#dt-chardata
|
|
549
|
-
PREDEFINED = {
|
|
550
|
-
38 => '&', # ampersand
|
|
551
|
-
60 => '<', # left angle bracket
|
|
552
|
-
62 => '>'} # right angle bracket
|
|
553
|
-
|
|
554
|
-
# http://www.w3.org/TR/REC-xml/#charsets
|
|
555
|
-
VALID = [[0x9, 0xA, 0xD], (0x20..0xD7FF),
|
|
556
|
-
(0xE000..0xFFFD), (0x10000..0x10FFFF)]
|
|
557
|
-
end
|
|
558
|
-
|
|
559
|
-
class Fixnum
|
|
560
|
-
# xml escaped version of chr
|
|
561
|
-
def xchr
|
|
562
|
-
n = XChar::CP1252[self] || self
|
|
563
|
-
n = 42 unless XChar::VALID.find {|range| range.include? n}
|
|
564
|
-
XChar::PREDEFINED[n] or (n<128 ? n.chr : "&##{n};")
|
|
565
|
-
end
|
|
566
|
-
end
|
|
567
|
-
|
|
568
|
-
class String
|
|
569
|
-
alias :old_index :index
|
|
570
|
-
def to_xs
|
|
571
|
-
unpack('U*').map {|n| n.xchr}.join # ASCII, UTF-8
|
|
572
|
-
rescue
|
|
573
|
-
unpack('C*').map {|n| n.xchr}.join # ISO-8859-1, WIN-1252
|
|
574
|
-
end
|
|
575
|
-
end
|
|
576
|
-
|
|
577
|
-
class BetterSGMLParserError < Exception; end;
|
|
578
|
-
class BetterSGMLParser < HTML::SGMLParser
|
|
579
|
-
# Replaced Tagfind and Charref Regexps with the ones in feedparser.py
|
|
580
|
-
# This makes things work.
|
|
581
|
-
Interesting = /[&<]/u
|
|
582
|
-
Incomplete = Regexp.compile('&([a-zA-Z][a-zA-Z0-9]*|#[0-9]*)?|' +
|
|
583
|
-
'<([a-zA-Z][^<>]*|/([a-zA-Z][^<>]*)?|' +
|
|
584
|
-
'![^<>]*)?', 64) # 64 is the unicode flag
|
|
585
|
-
|
|
586
|
-
Entityref = /&([a-zA-Z][-.a-zA-Z0-9]*)[^-.a-zA-Z0-9]/u
|
|
587
|
-
Charref = /&#(x?[0-9A-Fa-f]+)[^0-9A-Fa-f]/u
|
|
588
|
-
|
|
589
|
-
Shorttagopen = /'<[a-zA-Z][-.a-zA-Z0-9]*/u
|
|
590
|
-
Shorttag = /'<([a-zA-Z][-.a-zA-Z0-9]*)\/([^\/]*)\//u
|
|
591
|
-
Endtagopen = /<\//u # Matching the Python SGMLParser
|
|
592
|
-
Endbracket = /[<>]/u
|
|
593
|
-
Declopen = /<!/u
|
|
594
|
-
Piopenbegin = /^<\?/u
|
|
595
|
-
Piclose = />/u
|
|
596
|
-
|
|
597
|
-
Commentopen = /<!--/u
|
|
598
|
-
Commentclose = /--\s*>/u
|
|
599
|
-
Tagfind = /[a-zA-Z][-_.:a-zA-Z0-9]*/u
|
|
600
|
-
Attrfind = Regexp.compile('\s*([a-zA-Z_][-:.a-zA-Z_0-9]*)(\s*=\s*'+
|
|
601
|
-
'(\'[^\']*\'|"[^"]*"|[\]\[\-a-zA-Z0-9./,:;+*%?!&$\(\)_#=~\'"@]*))?',
|
|
602
|
-
64)
|
|
603
|
-
Endtagfind = /\s*\/\s*>/u
|
|
604
|
-
def initialize(verbose=false)
|
|
605
|
-
super(verbose)
|
|
606
|
-
end
|
|
607
|
-
def feed(*args)
|
|
608
|
-
super(*args)
|
|
609
|
-
end
|
|
610
|
-
|
|
611
|
-
def goahead(_end)
|
|
612
|
-
rawdata = @rawdata # woo, utf-8 magic
|
|
613
|
-
i = 0
|
|
614
|
-
n = rawdata.length
|
|
615
|
-
while i < n
|
|
616
|
-
if @nomoretags
|
|
617
|
-
# handle_data_range does nothing more than set a "Range" that is never used. wtf?
|
|
618
|
-
handle_data(rawdata[i...n]) # i...n means "range from i to n not including n"
|
|
619
|
-
i = n
|
|
620
|
-
break
|
|
621
|
-
end
|
|
622
|
-
j = rawdata.index(Interesting, i)
|
|
623
|
-
j = n unless j
|
|
624
|
-
handle_data(rawdata[i...j]) if i < j
|
|
625
|
-
i = j
|
|
626
|
-
break if (i == n)
|
|
627
|
-
if rawdata[i..i] == '<' # equivalent to rawdata[i..i] == '<' # Yeah, ugly.
|
|
628
|
-
if rawdata.index(Starttagopen,i) == i
|
|
629
|
-
if @literal
|
|
630
|
-
handle_data(rawdata[i..i])
|
|
631
|
-
i = i+1
|
|
632
|
-
next
|
|
633
|
-
end
|
|
634
|
-
k = parse_starttag(i)
|
|
635
|
-
break unless k
|
|
636
|
-
i = k
|
|
637
|
-
next
|
|
638
|
-
end
|
|
639
|
-
if rawdata.index(Endtagopen,i) == i #Don't use Endtagopen
|
|
640
|
-
k = parse_endtag(i)
|
|
641
|
-
break unless k
|
|
642
|
-
i = k
|
|
643
|
-
@literal = false
|
|
644
|
-
next
|
|
645
|
-
end
|
|
646
|
-
if @literal
|
|
647
|
-
if n > (i+1)
|
|
648
|
-
handle_data("<")
|
|
649
|
-
i = i+1
|
|
650
|
-
else
|
|
651
|
-
#incomplete
|
|
652
|
-
break
|
|
653
|
-
end
|
|
654
|
-
next
|
|
655
|
-
end
|
|
656
|
-
if rawdata.index(Commentopen,i) == i
|
|
657
|
-
k = parse_comment(i)
|
|
658
|
-
break unless k
|
|
659
|
-
i = k
|
|
660
|
-
next
|
|
661
|
-
end
|
|
662
|
-
if rawdata.index(Piopenbegin,i) == i # Like Piopen but must be at beginning of rawdata
|
|
663
|
-
k = parse_pi(i)
|
|
664
|
-
break unless k
|
|
665
|
-
i += k
|
|
666
|
-
next
|
|
667
|
-
end
|
|
668
|
-
if rawdata.index(Declopen,i) == i
|
|
669
|
-
# This is some sort of declaration; in "HTML as
|
|
670
|
-
# deployed," this should only be the document type
|
|
671
|
-
# declaration ("<!DOCTYPE html...>").
|
|
672
|
-
k = parse_declaration(i)
|
|
673
|
-
break unless k
|
|
674
|
-
i = k
|
|
675
|
-
next
|
|
676
|
-
end
|
|
677
|
-
elsif rawdata[i..i] == '&'
|
|
678
|
-
if @literal # FIXME BUGME SGMLParser totally does not check this. Bug it.
|
|
679
|
-
handle_data(rawdata[i..i])
|
|
680
|
-
i += 1
|
|
681
|
-
next
|
|
682
|
-
end
|
|
683
|
-
|
|
684
|
-
# the Char must come first as its #=~ method is the only one that is UTF-8 safe
|
|
685
|
-
ni,match = index_match(rawdata, Charref, i)
|
|
686
|
-
if ni and ni == i # See? Ugly
|
|
687
|
-
handle_charref(match[1]) # $1 is just the first group we captured (with parentheses)
|
|
688
|
-
i += match[0].length # $& is the "all" of the match.. it includes the full match we looked for not just the stuff we put parentheses around to capture.
|
|
689
|
-
i -= 1 unless rawdata[i-1..i-1] == ";"
|
|
690
|
-
next
|
|
691
|
-
end
|
|
692
|
-
ni,match = index_match(rawdata, Entityref, i)
|
|
693
|
-
if ni and ni == i
|
|
694
|
-
handle_entityref(match[1])
|
|
695
|
-
i += match[0].length
|
|
696
|
-
i -= 1 unless rawdata[i-1..i-1] == ";"
|
|
697
|
-
next
|
|
698
|
-
end
|
|
699
|
-
else
|
|
700
|
-
error('neither < nor & ??')
|
|
701
|
-
end
|
|
702
|
-
# We get here only if incomplete matches but
|
|
703
|
-
# nothing else
|
|
704
|
-
ni,match = index_match(rawdata,Incomplete,i)
|
|
705
|
-
unless ni and ni == 0
|
|
706
|
-
handle_data(rawdata[i...i+1]) # str[i...i+1] == str[i..i]
|
|
707
|
-
i += 1
|
|
708
|
-
next
|
|
709
|
-
end
|
|
710
|
-
j = ni + match[0].length
|
|
711
|
-
break if j == n # Really incomplete
|
|
712
|
-
handle_data(rawdata[i...j])
|
|
713
|
-
i = j
|
|
714
|
-
end # end while
|
|
715
|
-
|
|
716
|
-
if _end and i < n
|
|
717
|
-
handle_data(rawdata[i...n])
|
|
718
|
-
i = n
|
|
719
|
-
end
|
|
720
|
-
|
|
721
|
-
@rawdata = rawdata[i..-1]
|
|
722
|
-
# @offset += i # FIXME BUGME another unused variable in SGMLParser?
|
|
723
|
-
end
|
|
724
|
-
|
|
725
|
-
|
|
726
|
-
# Internal -- parse processing instr, return length or -1 if not terminated
|
|
727
|
-
def parse_pi(i)
|
|
728
|
-
rawdata = @rawdata
|
|
729
|
-
if rawdata[i...i+2] != '<?'
|
|
730
|
-
error("unexpected call to parse_pi()")
|
|
731
|
-
end
|
|
732
|
-
ni,match = index_match(rawdata,Piclose,i+2)
|
|
733
|
-
return nil unless match
|
|
734
|
-
j = ni
|
|
735
|
-
handle_pi(rawdata[i+2...j])
|
|
736
|
-
j = (j + match[0].length)
|
|
737
|
-
return j-i
|
|
738
|
-
end
|
|
739
|
-
|
|
740
|
-
def parse_comment(i)
|
|
741
|
-
rawdata = @rawdata
|
|
742
|
-
if rawdata[i...i+4] != "<!--"
|
|
743
|
-
error("unexpected call to parse_comment()")
|
|
744
|
-
end
|
|
745
|
-
ni,match = index_match(rawdata, Commentclose,i)
|
|
746
|
-
return nil unless match
|
|
747
|
-
handle_comment(rawdata[i+4..(ni-1)])
|
|
748
|
-
return ni+match[0].length # Length from i to just past the closing comment tag
|
|
749
|
-
end
|
|
750
|
-
|
|
751
|
-
|
|
752
|
-
def parse_starttag(i)
|
|
753
|
-
@_starttag_text = nil
|
|
754
|
-
start_pos = i
|
|
755
|
-
rawdata = @rawdata
|
|
756
|
-
ni,match = index_match(rawdata,Shorttagopen,i)
|
|
757
|
-
if ni == i
|
|
758
|
-
# SGML shorthand: <tag/data/ == <tag>data</tag>
|
|
759
|
-
# XXX Can data contain &... (entity or char refs)?
|
|
760
|
-
# XXX Can data contain < or > (tag characters)?
|
|
761
|
-
# XXX Can there be whitespace before the first /?
|
|
762
|
-
k,match = index_match(rawdata,Shorttag,i)
|
|
763
|
-
return nil unless match
|
|
764
|
-
tag, data = match[1], match[2]
|
|
765
|
-
@_starttag_text = "<#{tag}/"
|
|
766
|
-
tag.downcase!
|
|
767
|
-
second_end = rawdata.index(Shorttagopen,k)
|
|
768
|
-
finish_shorttag(tag, data)
|
|
769
|
-
@_starttag_text = rawdata[start_pos...second_end+1]
|
|
770
|
-
return k
|
|
771
|
-
end
|
|
772
|
-
|
|
773
|
-
j = rawdata.index(Endbracket, i+1)
|
|
774
|
-
return nil unless j
|
|
775
|
-
attrsd = []
|
|
776
|
-
if rawdata[i...i+2] == '<>'
|
|
777
|
-
# SGML shorthand: <> == <last open tag seen>
|
|
778
|
-
k = j
|
|
779
|
-
tag = @lasttag
|
|
780
|
-
else
|
|
781
|
-
ni,match = index_match(rawdata,Tagfind,i+1)
|
|
782
|
-
unless match
|
|
783
|
-
error('unexpected call to parse_starttag')
|
|
784
|
-
end
|
|
785
|
-
k = ni+match[0].length+1
|
|
786
|
-
tag = match[0].downcase
|
|
787
|
-
@lasttag = tag
|
|
788
|
-
end
|
|
789
|
-
|
|
790
|
-
while k < j
|
|
791
|
-
break if rawdata.index(Endtagfind, k) == k
|
|
792
|
-
ni,match = index_match(rawdata,Attrfind,k)
|
|
793
|
-
break unless ni
|
|
794
|
-
matched_length = match[0].length
|
|
795
|
-
attrname, rest, attrvalue = match[1],match[2],match[3]
|
|
796
|
-
if rest.nil? or rest.empty?
|
|
797
|
-
attrvalue = '' # was: = attrname # Why the change?
|
|
798
|
-
elsif [?',?'] == [attrvalue[0..0], attrvalue[-1..-1]] or [?",?"] == [attrvalue[0],attrvalue[-1]]
|
|
799
|
-
attrvalue = attrvalue[1...-1]
|
|
800
|
-
end
|
|
801
|
-
attrsd << [attrname.downcase, attrvalue]
|
|
802
|
-
k += matched_length
|
|
803
|
-
end
|
|
804
|
-
if rawdata[j..j] == ">"
|
|
805
|
-
j += 1
|
|
806
|
-
end
|
|
807
|
-
@_starttag_text = rawdata[start_pos...j]
|
|
808
|
-
finish_starttag(tag, attrsd)
|
|
809
|
-
return j
|
|
810
|
-
end
|
|
811
|
-
|
|
812
|
-
def parse_endtag(i)
|
|
813
|
-
rawdata = @rawdata
|
|
814
|
-
j, match = index_match(rawdata, /[<>]/,i+1)
|
|
815
|
-
return nil unless j
|
|
816
|
-
tag = rawdata[i+2...j].strip.downcase
|
|
817
|
-
if rawdata[j..j] == ">"
|
|
818
|
-
j += 1
|
|
819
|
-
end
|
|
820
|
-
finish_endtag(tag)
|
|
821
|
-
return j
|
|
822
|
-
end
|
|
823
|
-
|
|
824
|
-
def output
|
|
825
|
-
# Return processed HTML as a single string
|
|
826
|
-
return @pieces.map{|p| p.to_s}.join
|
|
827
|
-
end
|
|
828
|
-
|
|
829
|
-
def error(message)
|
|
830
|
-
raise BetterSGMLParserError.new(message)
|
|
831
|
-
end
|
|
832
|
-
def handle_pi(text)
|
|
833
|
-
end
|
|
834
|
-
def handle_decl(text)
|
|
835
|
-
end
|
|
836
|
-
end
|
|
837
|
-
|
|
838
|
-
# Add some helper methods to make AttributeList (all of those damn attrs
|
|
839
|
-
# and attrsD used by StrictFeedParser) act more like a Hash.
|
|
840
|
-
# NOTE AttributeList is still Read-Only (AFAICT).
|
|
841
|
-
# Monkey patching is terrible, and I have an addiction.
|
|
842
|
-
module XML
|
|
843
|
-
module SAX
|
|
844
|
-
module AttributeList # in xml/sax.rb
|
|
845
|
-
def [](key)
|
|
846
|
-
getValue(key)
|
|
847
|
-
end
|
|
848
|
-
|
|
849
|
-
def each(&blk)
|
|
850
|
-
(0...getLength).each{|pos| yield [getName(pos), getValue(pos)]}
|
|
851
|
-
end
|
|
852
|
-
|
|
853
|
-
def each_key(&blk)
|
|
854
|
-
(0...getLength).each{|pos| yield getName(pos) }
|
|
855
|
-
end
|
|
856
|
-
|
|
857
|
-
def each_value(&blk)
|
|
858
|
-
(0...getLength).each{|pos| yield getValue(pos) }
|
|
859
|
-
end
|
|
860
|
-
|
|
861
|
-
def to_a # Rather use collect? grep for to_a.collect
|
|
862
|
-
l = []
|
|
863
|
-
each{|k,v| l << [k,v]}
|
|
864
|
-
return l
|
|
865
|
-
end
|
|
866
|
-
|
|
867
|
-
def to_s
|
|
868
|
-
l = []
|
|
869
|
-
each{|k,v| l << "#{k} => #{v}"}
|
|
870
|
-
"{ "+l.join(", ")+" }"
|
|
871
|
-
end
|
|
872
|
-
end
|
|
873
|
-
end
|
|
874
|
-
end
|
|
875
|
-
|
|
876
|
-
# This used to be based on Michael Moen's Hpricot#scrub, but that seems to
|
|
877
|
-
# have only been part of its evolution. Hpricot#scrub is cool code, though.
|
|
878
|
-
# http://underpantsgnome.com/2007/01/20/hpricot-scrub
|
|
879
|
-
module Hpricot
|
|
880
|
-
Acceptable_Elements = ['a', 'abbr', 'acronym', 'address', 'area', 'b',
|
|
881
|
-
'big', 'blockquote', 'br', 'button', 'caption', 'center', 'cite',
|
|
882
|
-
'code', 'col', 'colgroup', 'dd', 'del', 'dfn', 'dir', 'div', 'dl', 'dt',
|
|
883
|
-
'em', 'fieldset', 'font', 'form', 'h1', 'h2', 'h3', 'h4', 'h5', 'h6',
|
|
884
|
-
'hr', 'i', 'img', 'input', 'ins', 'kbd', 'label', 'legend', 'li', 'map',
|
|
885
|
-
'menu', 'ol', 'optgroup', 'option', 'p', 'pre', 'q', 's', 'samp',
|
|
886
|
-
'select', 'small', 'span', 'strike', 'strong', 'sub', 'sup', 'table',
|
|
887
|
-
'tbody', 'td', 'textarea', 'tfoot', 'th', 'thead', 'tr', 'tt', 'u',
|
|
888
|
-
'ul', 'var'
|
|
889
|
-
]
|
|
890
|
-
|
|
891
|
-
Acceptable_Attributes = ['abbr', 'accept', 'accept-charset', 'accesskey',
|
|
892
|
-
'action', 'align', 'alt', 'axis', 'border', 'cellpadding',
|
|
893
|
-
'cellspacing', 'char', 'charoff', 'charset', 'checked', 'cite', 'class',
|
|
894
|
-
'clear', 'cols', 'colspan', 'color', 'compact', 'coords', 'datetime',
|
|
895
|
-
'dir', 'disabled', 'enctype', 'for', 'frame', 'headers', 'height',
|
|
896
|
-
'href', 'hreflang', 'hspace', 'id', 'ismap', 'label', 'lang',
|
|
897
|
-
'longdesc', 'maxlength', 'media', 'method', 'multiple', 'name',
|
|
898
|
-
'nohref', 'noshade', 'nowrap', 'prompt', 'readonly', 'rel', 'rev',
|
|
899
|
-
'rows', 'rowspan', 'rules', 'scope', 'selected', 'shape', 'size',
|
|
900
|
-
'span', 'src', 'start', 'summary', 'tabindex', 'target', 'title',
|
|
901
|
-
'type', 'usemap', 'valign', 'value', 'vspace', 'width', 'xml:lang'
|
|
902
|
-
]
|
|
903
|
-
|
|
904
|
-
Unacceptable_Elements_With_End_Tag = ['script', 'applet']
|
|
905
|
-
|
|
906
|
-
Acceptable_Css_Properties = ['azimuth', 'background-color',
|
|
907
|
-
'border-bottom-color', 'border-collapse', 'border-color',
|
|
908
|
-
'border-left-color', 'border-right-color', 'border-top-color', 'clear',
|
|
909
|
-
'color', 'cursor', 'direction', 'display', 'elevation', 'float', 'font',
|
|
910
|
-
'font-family', 'font-size', 'font-style', 'font-variant', 'font-weight',
|
|
911
|
-
'height', 'letter-spacing', 'line-height', 'overflow', 'pause',
|
|
912
|
-
'pause-after', 'pause-before', 'pitch', 'pitch-range', 'richness',
|
|
913
|
-
'speak', 'speak-header', 'speak-numeral', 'speak-punctuation',
|
|
914
|
-
'speech-rate', 'stress', 'text-align', 'text-decoration', 'text-indent',
|
|
915
|
-
'unicode-bidi', 'vertical-align', 'voice-family', 'volume',
|
|
916
|
-
'white-space', 'width'
|
|
917
|
-
]
|
|
918
|
-
|
|
919
|
-
# survey of common keywords found in feeds
|
|
920
|
-
Acceptable_Css_Keywords = ['auto', 'aqua', 'black', 'block', 'blue',
|
|
921
|
-
'bold', 'both', 'bottom', 'brown', 'center', 'collapse', 'dashed',
|
|
922
|
-
'dotted', 'fuchsia', 'gray', 'green', '!important', 'italic', 'left',
|
|
923
|
-
'lime', 'maroon', 'medium', 'none', 'navy', 'normal', 'nowrap', 'olive',
|
|
924
|
-
'pointer', 'purple', 'red', 'right', 'solid', 'silver', 'teal', 'top',
|
|
925
|
-
'transparent', 'underline', 'white', 'yellow'
|
|
926
|
-
]
|
|
927
|
-
|
|
928
|
-
Mathml_Elements = ['maction', 'math', 'merror', 'mfrac', 'mi',
|
|
929
|
-
'mmultiscripts', 'mn', 'mo', 'mover', 'mpadded', 'mphantom',
|
|
930
|
-
'mprescripts', 'mroot', 'mrow', 'mspace', 'msqrt', 'mstyle', 'msub',
|
|
931
|
-
'msubsup', 'msup', 'mtable', 'mtd', 'mtext', 'mtr', 'munder',
|
|
932
|
-
'munderover', 'none'
|
|
933
|
-
]
|
|
934
|
-
|
|
935
|
-
Mathml_Attributes = ['actiontype', 'align', 'columnalign', 'columnalign',
|
|
936
|
-
'columnalign', 'columnlines', 'columnspacing', 'columnspan', 'depth',
|
|
937
|
-
'display', 'displaystyle', 'equalcolumns', 'equalrows', 'fence',
|
|
938
|
-
'fontstyle', 'fontweight', 'frame', 'height', 'linethickness', 'lspace',
|
|
939
|
-
'mathbackground', 'mathcolor', 'mathvariant', 'mathvariant', 'maxsize',
|
|
940
|
-
'minsize', 'other', 'rowalign', 'rowalign', 'rowalign', 'rowlines',
|
|
941
|
-
'rowspacing', 'rowspan', 'rspace', 'scriptlevel', 'selection',
|
|
942
|
-
'separator', 'stretchy', 'width', 'width', 'xlink:href', 'xlink:show',
|
|
943
|
-
'xlink:type', 'xmlns', 'xmlns:xlink'
|
|
944
|
-
]
|
|
945
|
-
|
|
946
|
-
# svgtiny - foreignObject + linearGradient + radialGradient + stop
|
|
947
|
-
Svg_Elements = ['a', 'animate', 'animateColor', 'animateMotion',
|
|
948
|
-
'animateTransform', 'circle', 'defs', 'desc', 'ellipse', 'font-face',
|
|
949
|
-
'font-face-name', 'font-face-src', 'g', 'glyph', 'hkern', 'image',
|
|
950
|
-
'linearGradient', 'line', 'metadata', 'missing-glyph', 'mpath', 'path',
|
|
951
|
-
'polygon', 'polyline', 'radialGradient', 'rect', 'set', 'stop', 'svg',
|
|
952
|
-
'switch', 'text', 'title', 'use'
|
|
953
|
-
]
|
|
954
|
-
|
|
955
|
-
# svgtiny + class + opacity + offset + xmlns + xmlns:xlink
|
|
956
|
-
Svg_Attributes = ['accent-height', 'accumulate', 'additive', 'alphabetic',
|
|
957
|
-
'arabic-form', 'ascent', 'attributeName', 'attributeType',
|
|
958
|
-
'baseProfile', 'bbox', 'begin', 'by', 'calcMode', 'cap-height',
|
|
959
|
-
'class', 'color', 'color-rendering', 'content', 'cx', 'cy', 'd',
|
|
960
|
-
'descent', 'display', 'dur', 'end', 'fill', 'fill-rule', 'font-family',
|
|
961
|
-
'font-size', 'font-stretch', 'font-style', 'font-variant',
|
|
962
|
-
'font-weight', 'from', 'fx', 'fy', 'g1', 'g2', 'glyph-name',
|
|
963
|
-
'gradientUnits', 'hanging', 'height', 'horiz-adv-x', 'horiz-origin-x',
|
|
964
|
-
'id', 'ideographic', 'k', 'keyPoints', 'keySplines', 'keyTimes',
|
|
965
|
-
'lang', 'mathematical', 'max', 'min', 'name', 'offset', 'opacity',
|
|
966
|
-
'origin', 'overline-position', 'overline-thickness', 'panose-1',
|
|
967
|
-
'path', 'pathLength', 'points', 'preserveAspectRatio', 'r',
|
|
968
|
-
'repeatCount', 'repeatDur', 'requiredExtensions', 'requiredFeatures',
|
|
969
|
-
'restart', 'rotate', 'rx', 'ry', 'slope', 'stemh', 'stemv',
|
|
970
|
-
'stop-color', 'stop-opacity', 'strikethrough-position',
|
|
971
|
-
'strikethrough-thickness', 'stroke', 'stroke-dasharray',
|
|
972
|
-
'stroke-dashoffset', 'stroke-linecap', 'stroke-linejoin',
|
|
973
|
-
'stroke-miterlimit', 'stroke-width', 'systemLanguage', 'target',
|
|
974
|
-
'text-anchor', 'to', 'transform', 'type', 'u1', 'u2',
|
|
975
|
-
'underline-position', 'underline-thickness', 'unicode',
|
|
976
|
-
'unicode-range', 'units-per-em', 'values', 'version', 'viewBox',
|
|
977
|
-
'visibility', 'width', 'widths', 'x', 'x-height', 'x1', 'x2',
|
|
978
|
-
'xlink:actuate', 'xlink:arcrole', 'xlink:href', 'xlink:role',
|
|
979
|
-
'xlink:show', 'xlink:title', 'xlink:type', 'xml:base', 'xml:lang',
|
|
980
|
-
'xml:space', 'xmlns', 'xmlns:xlink', 'y', 'y1', 'y2', 'zoomAndPan'
|
|
981
|
-
]
|
|
982
|
-
|
|
983
|
-
Svg_Attr_Map = nil
|
|
984
|
-
Svg_Elem_Map = nil
|
|
985
|
-
|
|
986
|
-
Acceptable_Svg_Properties = [ 'fill', 'fill-opacity', 'fill-rule',
|
|
987
|
-
'stroke', 'stroke-width', 'stroke-linecap', 'stroke-linejoin',
|
|
988
|
-
'stroke-opacity'
|
|
989
|
-
]
|
|
990
|
-
|
|
991
|
-
unless $compatible
|
|
992
|
-
@@acceptable_tag_specific_attributes = {}
|
|
993
|
-
@@mathml_elements.each{|e| @@acceptable_tag_specific_attributes[e] = @@mathml_attributes }
|
|
994
|
-
@@svg_elements.each{|e| @@acceptable_tag_specific_attributes[e] = @@svg_attributes }
|
|
995
|
-
end
|
|
996
|
-
|
|
997
|
-
class Elements
|
|
998
|
-
def strip_attributes(safe=[])
|
|
999
|
-
each { |x| x.strip_attributes(safe) }
|
|
1000
|
-
end
|
|
1001
|
-
|
|
1002
|
-
def strip_style(ok_props=[], ok_keywords=[]) # NOTE unused so far.
|
|
1003
|
-
each { |x| x.strip_style(ok_props, ok_keywords) }
|
|
1004
|
-
end
|
|
1005
|
-
end
|
|
42
|
+
$LOAD_PATH << File.expand_path(File.dirname(__FILE__))
|
|
43
|
+
require 'rfeedparser/forgiving_uri'
|
|
44
|
+
require 'rfeedparser/aliases'
|
|
45
|
+
require 'rfeedparser/encoding_helpers'
|
|
46
|
+
require 'rfeedparser/better_sgmlparser'
|
|
47
|
+
require 'rfeedparser/better_attributelist'
|
|
48
|
+
require 'rfeedparser/scrub'
|
|
49
|
+
require 'rfeedparser/time_helpers'
|
|
50
|
+
require 'rfeedparser/feedparserdict'
|
|
51
|
+
require 'rfeedparser/parser_mixin'
|
|
52
|
+
require 'rfeedparser/parsers'
|
|
53
|
+
require 'rfeedparser/markup_helpers'
|
|
1006
54
|
|
|
1007
|
-
|
|
1008
|
-
def strip_attributes(foo)
|
|
1009
|
-
end
|
|
1010
|
-
end
|
|
1011
|
-
class Comment
|
|
1012
|
-
def strip_attributes(foo)
|
|
1013
|
-
end
|
|
1014
|
-
end
|
|
1015
|
-
class BogusETag
|
|
1016
|
-
def strip_attributes(foo)
|
|
1017
|
-
end
|
|
1018
|
-
end
|
|
55
|
+
include FeedParserUtilities
|
|
1019
56
|
|
|
1020
|
-
class Elem
|
|
1021
|
-
def strip_attributes
|
|
1022
|
-
unless attributes.nil?
|
|
1023
|
-
attributes.each do |atr|
|
|
1024
|
-
unless Acceptable_Attributes.include?atr[0]
|
|
1025
|
-
remove_attribute(atr[0])
|
|
1026
|
-
end
|
|
1027
|
-
end
|
|
1028
|
-
end
|
|
1029
|
-
end
|
|
1030
|
-
end
|
|
1031
|
-
end
|
|
1032
57
|
|
|
1033
58
|
module FeedParser
|
|
1034
|
-
Version = "0.9.
|
|
59
|
+
Version = "0.9.91"
|
|
1035
60
|
|
|
1036
61
|
License = """Copyright (c) 2002-2006, Mark Pilgrim, All rights reserved.
|
|
1037
62
|
|
|
@@ -1059,10 +84,10 @@ POSSIBILITY OF SUCH DAMAGE."""
|
|
|
1059
84
|
Author = "Jeff Hodges <http://somethingsimilar.com>"
|
|
1060
85
|
Copyright_Holder = "Mark Pilgrim <http://diveintomark.org/>"
|
|
1061
86
|
Contributors = [ "Jason Diamond <http://injektilo.org/>",
|
|
1062
|
-
|
|
1063
|
-
|
|
1064
|
-
|
|
1065
|
-
|
|
87
|
+
"John Beimler <http://john.beimler.org/>",
|
|
88
|
+
"Fazal Majid <http://www.majid.info/mylos/weblog/>",
|
|
89
|
+
"Aaron Swartz <http://aaronsw.com/>",
|
|
90
|
+
"Kevin Marks <http://epeus.blogspot.com/>"
|
|
1066
91
|
]
|
|
1067
92
|
# HTTP "User-Agent" header to send to servers when downloading feeds.
|
|
1068
93
|
# If you are embedding feedparser in a larger application, you should
|
|
@@ -1115,2207 +140,27 @@ POSSIBILITY OF SUCH DAMAGE."""
|
|
|
1115
140
|
'cdf' => 'CDF',
|
|
1116
141
|
'hotrss' => 'Hot RSS'
|
|
1117
142
|
}
|
|
1118
|
-
|
|
1119
|
-
=
|
|
1120
|
-
|
|
1121
|
-
|
|
1122
|
-
|
|
1123
|
-
|
|
1124
|
-
|
|
1125
|
-
|
|
1126
|
-
|
|
1127
|
-
|
|
1128
|
-
|
|
1129
|
-
|
|
1130
|
-
with possible "correct" attributes in the its values. the #[] and #[]=
|
|
1131
|
-
methods check with keymaps to see what attribute the developer "really
|
|
1132
|
-
means" if they've asked for one which happens to be in @@keymap's keys.
|
|
1133
|
-
=end
|
|
1134
|
-
@@keymap = {'channel' => 'feed',
|
|
1135
|
-
'items' => 'entries',
|
|
1136
|
-
'guid' => 'id',
|
|
1137
|
-
'date' => 'updated',
|
|
1138
|
-
'date_parsed' => 'updated_parsed',
|
|
1139
|
-
'description' => ['subtitle', 'summary'],
|
|
1140
|
-
'url' => ['href'],
|
|
1141
|
-
'modified' => 'updated',
|
|
1142
|
-
'modified_parsed' => 'updated_parsed',
|
|
1143
|
-
'issued' => 'published',
|
|
1144
|
-
'issued_parsed' => 'published_parsed',
|
|
1145
|
-
'copyright' => 'rights',
|
|
1146
|
-
'copyright_detail' => 'rights_detail',
|
|
1147
|
-
'tagline' => 'subtitle',
|
|
1148
|
-
'tagline_detail' => 'subtitle_detail'}
|
|
1149
|
-
|
|
1150
|
-
def entries # Apparently, Hash has an entries method! That blew a good 3 hours or more of my time
|
|
1151
|
-
return self['entries']
|
|
1152
|
-
end
|
|
1153
|
-
|
|
1154
|
-
# We could include the [] rewrite in new using Hash.new's fancy pants block thing
|
|
1155
|
-
# but we'd still have to overwrite []= and such.
|
|
1156
|
-
# I'm going to make it easy to turn lists of pairs into FeedParserDicts's though.
|
|
1157
|
-
def initialize(pairs=nil)
|
|
1158
|
-
if pairs.class == Array and pairs[0].class == Array and pairs[0].length == 2
|
|
1159
|
-
pairs.each do |l|
|
|
1160
|
-
k,v = l
|
|
1161
|
-
self[k] = v
|
|
1162
|
-
end
|
|
1163
|
-
elsif pairs.class == Hash
|
|
1164
|
-
self.merge!(pairs)
|
|
1165
|
-
end
|
|
1166
|
-
end
|
|
1167
|
-
|
|
1168
|
-
def [](key)
|
|
1169
|
-
if key == 'category'
|
|
1170
|
-
return self['tags'][0]['term']
|
|
1171
|
-
end
|
|
1172
|
-
if key == 'categories'
|
|
1173
|
-
return self['tags'].collect{|tag| [tag['scheme'],tag['term']]}
|
|
1174
|
-
end
|
|
1175
|
-
realkey = @@keymap[key] || key
|
|
1176
|
-
if realkey.class == Array
|
|
1177
|
-
realkey.each{ |key| return self[key] if has_key?key }
|
|
1178
|
-
end
|
|
1179
|
-
# Note that the original key is preferred over the realkey we (might
|
|
1180
|
-
# have) found in @@keymap
|
|
1181
|
-
if has_key?(key)
|
|
1182
|
-
return super(key)
|
|
1183
|
-
end
|
|
1184
|
-
return super(realkey)
|
|
1185
|
-
end
|
|
1186
|
-
|
|
1187
|
-
def []=(key,value)
|
|
1188
|
-
if @@keymap.key?key
|
|
1189
|
-
key = @@keymap[key]
|
|
1190
|
-
if key.class == Array
|
|
1191
|
-
key = key[0]
|
|
1192
|
-
end
|
|
1193
|
-
end
|
|
1194
|
-
super(key,value)
|
|
1195
|
-
end
|
|
1196
|
-
|
|
1197
|
-
def method_missing(msym, *args)
|
|
1198
|
-
methodname = msym.to_s
|
|
1199
|
-
if methodname[-1] == '='
|
|
1200
|
-
return self[methodname[0..-2]] = args[0]
|
|
1201
|
-
elsif methodname[-1] != '!' and methodname[-1] != '?' and methodname[0] != "_" # FIXME implement with private
|
|
1202
|
-
return self[methodname]
|
|
1203
|
-
else
|
|
1204
|
-
raise NoMethodError, "whoops, we don't know about the attribute or method called `#{methodname}' for #{self}:#{self.class}"
|
|
1205
|
-
end
|
|
1206
|
-
end
|
|
1207
|
-
end
|
|
1208
|
-
|
|
1209
|
-
|
|
1210
|
-
|
|
1211
|
-
|
|
1212
|
-
module FeedParserMixin
|
|
1213
|
-
attr_accessor :feeddata, :version, :namespacesInUse, :date_handlers
|
|
1214
|
-
|
|
1215
|
-
def startup(baseuri=nil, baselang=nil, encoding='utf-8')
|
|
1216
|
-
$stderr << "initializing FeedParser\n" if $debug
|
|
1217
|
-
|
|
1218
|
-
@namespaces = {'' => '',
|
|
1219
|
-
'http://backend.userland.com/rss' => '',
|
|
1220
|
-
'http://blogs.law.harvard.edu/tech/rss' => '',
|
|
1221
|
-
'http://purl.org/rss/1.0/' => '',
|
|
1222
|
-
'http://my.netscape.com/rdf/simple/0.9/' => '',
|
|
1223
|
-
'http://example.com/newformat#' => '',
|
|
1224
|
-
'http://example.com/necho' => '',
|
|
1225
|
-
'http://purl.org/echo/' => '',
|
|
1226
|
-
'uri/of/echo/namespace#' => '',
|
|
1227
|
-
'http://purl.org/pie/' => '',
|
|
1228
|
-
'http://purl.org/atom/ns#' => '',
|
|
1229
|
-
'http://www.w3.org/2005/Atom' => '',
|
|
1230
|
-
'http://purl.org/rss/1.0/modules/rss091#' => '',
|
|
1231
|
-
'http://webns.net/mvcb/' => 'admin',
|
|
1232
|
-
'http://purl.org/rss/1.0/modules/aggregation/' => 'ag',
|
|
1233
|
-
'http://purl.org/rss/1.0/modules/annotate/' => 'annotate',
|
|
1234
|
-
'http://media.tangent.org/rss/1.0/' => 'audio',
|
|
1235
|
-
'http://backend.userland.com/blogChannelModule' => 'blogChannel',
|
|
1236
|
-
'http://web.resource.org/cc/' => 'cc',
|
|
1237
|
-
'http://backend.userland.com/creativeCommonsRssModule' => 'creativeCommons',
|
|
1238
|
-
'http://purl.org/rss/1.0/modules/company' => 'co',
|
|
1239
|
-
'http://purl.org/rss/1.0/modules/content/' => 'content',
|
|
1240
|
-
'http://my.theinfo.org/changed/1.0/rss/' => 'cp',
|
|
1241
|
-
'http://purl.org/dc/elements/1.1/' => 'dc',
|
|
1242
|
-
'http://purl.org/dc/terms/' => 'dcterms',
|
|
1243
|
-
'http://purl.org/rss/1.0/modules/email/' => 'email',
|
|
1244
|
-
'http://purl.org/rss/1.0/modules/event/' => 'ev',
|
|
1245
|
-
'http://rssnamespace.org/feedburner/ext/1.0' => 'feedburner',
|
|
1246
|
-
'http://freshmeat.net/rss/fm/' => 'fm',
|
|
1247
|
-
'http://xmlns.com/foaf/0.1/' => 'foaf',
|
|
1248
|
-
'http://www.w3.org/2003/01/geo/wgs84_pos#' => 'geo',
|
|
1249
|
-
'http://postneo.com/icbm/' => 'icbm',
|
|
1250
|
-
'http://purl.org/rss/1.0/modules/image/' => 'image',
|
|
1251
|
-
'http://www.itunes.com/DTDs/PodCast-1.0.dtd' => 'itunes',
|
|
1252
|
-
'http://example.com/DTDs/PodCast-1.0.dtd' => 'itunes',
|
|
1253
|
-
'http://purl.org/rss/1.0/modules/link/' => 'l',
|
|
1254
|
-
'http://search.yahoo.com/mrss' => 'media',
|
|
1255
|
-
'http://madskills.com/public/xml/rss/module/pingback/' => 'pingback',
|
|
1256
|
-
'http://prismstandard.org/namespaces/1.2/basic/' => 'prism',
|
|
1257
|
-
'http://www.w3.org/1999/02/22-rdf-syntax-ns#' => 'rdf',
|
|
1258
|
-
'http://www.w3.org/2000/01/rdf-schema#' => 'rdfs',
|
|
1259
|
-
'http://purl.org/rss/1.0/modules/reference/' => 'ref',
|
|
1260
|
-
'http://purl.org/rss/1.0/modules/richequiv/' => 'reqv',
|
|
1261
|
-
'http://purl.org/rss/1.0/modules/search/' => 'search',
|
|
1262
|
-
'http://purl.org/rss/1.0/modules/slash/' => 'slash',
|
|
1263
|
-
'http://schemas.xmlsoap.org/soap/envelope/' => 'soap',
|
|
1264
|
-
'http://purl.org/rss/1.0/modules/servicestatus/' => 'ss',
|
|
1265
|
-
'http://hacks.benhammersley.com/rss/streaming/' => 'str',
|
|
1266
|
-
'http://purl.org/rss/1.0/modules/subscription/' => 'sub',
|
|
1267
|
-
'http://purl.org/rss/1.0/modules/syndication/' => 'sy',
|
|
1268
|
-
'http://purl.org/rss/1.0/modules/taxonomy/' => 'taxo',
|
|
1269
|
-
'http://purl.org/rss/1.0/modules/threading/' => 'thr',
|
|
1270
|
-
'http://purl.org/rss/1.0/modules/textinput/' => 'ti',
|
|
1271
|
-
'http://madskills.com/public/xml/rss/module/trackback/' =>'trackback',
|
|
1272
|
-
'http://wellformedweb.org/commentAPI/' => 'wfw',
|
|
1273
|
-
'http://purl.org/rss/1.0/modules/wiki/' => 'wiki',
|
|
1274
|
-
'http://www.w3.org/1999/xhtml' => 'xhtml',
|
|
1275
|
-
'http://www.w3.org/XML/1998/namespace' => 'xml',
|
|
1276
|
-
'http://www.w3.org/1999/xlink' => 'xlink',
|
|
1277
|
-
'http://schemas.pocketsoap.com/rss/myDescModule/' => 'szf'
|
|
1278
|
-
}
|
|
1279
|
-
@matchnamespaces = {}
|
|
1280
|
-
@namespaces.each do |l|
|
|
1281
|
-
@matchnamespaces[l[0].downcase] = l[1]
|
|
1282
|
-
end
|
|
1283
|
-
@can_be_relative_uri = ['link', 'id', 'wfw_comment', 'wfw_commentrss', 'docs', 'url', 'href', 'comments', 'license', 'icon', 'logo']
|
|
1284
|
-
@can_contain_relative_uris = ['content', 'title', 'summary', 'info', 'tagline', 'subtitle', 'copyright', 'rights', 'description']
|
|
1285
|
-
@can_contain_dangerous_markup = ['content', 'title', 'summary', 'info', 'tagline', 'subtitle', 'copyright', 'rights', 'description']
|
|
1286
|
-
@html_types = ['text/html', 'application/xhtml+xml']
|
|
1287
|
-
@feeddata = FeedParserDict.new # feed-level data
|
|
1288
|
-
@encoding = encoding # character encoding
|
|
1289
|
-
@entries = [] # list of entry-level data
|
|
1290
|
-
@version = '' # feed type/version see SUPPORTED_VERSIOSN
|
|
1291
|
-
@namespacesInUse = {} # hash of namespaces defined by the feed
|
|
1292
|
-
|
|
1293
|
-
# the following are used internall to track state;
|
|
1294
|
-
# this is really out of control and should be refactored
|
|
1295
|
-
@infeed = false
|
|
1296
|
-
@inentry = false
|
|
1297
|
-
@incontent = 0 # Yes, this needs to be zero until I work out popContent and pushContent
|
|
1298
|
-
@intextinput = false
|
|
1299
|
-
@inimage = false
|
|
1300
|
-
@inauthor = false
|
|
1301
|
-
@incontributor = false
|
|
1302
|
-
@inpublisher = false
|
|
1303
|
-
@insource = false
|
|
1304
|
-
@sourcedata = FeedParserDict.new
|
|
1305
|
-
@contentparams = FeedParserDict.new
|
|
1306
|
-
@summaryKey = nil
|
|
1307
|
-
@namespacemap = {}
|
|
1308
|
-
@elementstack = []
|
|
1309
|
-
@basestack = []
|
|
1310
|
-
@langstack = []
|
|
1311
|
-
@baseuri = baseuri || ''
|
|
1312
|
-
@lang = baselang || nil
|
|
1313
|
-
if baselang
|
|
1314
|
-
@feeddata['language'] = baselang.gsub('_','-')
|
|
1315
|
-
end
|
|
1316
|
-
@date_handlers = [:_parse_date_rfc822,
|
|
1317
|
-
:_parse_date_hungarian, :_parse_date_greek,:_parse_date_mssql,
|
|
1318
|
-
:_parse_date_nate,:_parse_date_onblog,:_parse_date_w3dtf,:_parse_date_iso8601
|
|
1319
|
-
]
|
|
1320
|
-
$stderr << "Leaving startup\n" if $debug # My addition
|
|
1321
|
-
end
|
|
1322
|
-
|
|
1323
|
-
def unknown_starttag(tag, attrsd)
|
|
1324
|
-
$stderr << "start #{tag} with #{attrsd}\n" if $debug
|
|
1325
|
-
# normalize attrs
|
|
1326
|
-
attrsD = {}
|
|
1327
|
-
attrsd = Hash[*attrsd.flatten] if attrsd.class == Array # Magic! Asterisk!
|
|
1328
|
-
# LooseFeedParser needs the above because SGMLParser sends attrs as a
|
|
1329
|
-
# list of lists (like [['type','text/html'],['mode','escaped']])
|
|
1330
|
-
|
|
1331
|
-
attrsd.each do |old_k,value|
|
|
1332
|
-
# There has to be a better, non-ugly way of doing this
|
|
1333
|
-
k = old_k.downcase # Downcase all keys
|
|
1334
|
-
attrsD[k] = value
|
|
1335
|
-
if ['rel','type'].include?value
|
|
1336
|
-
attrsD[k].downcase! # Downcase the value if the key is 'rel' or 'type'
|
|
1337
|
-
end
|
|
1338
|
-
end
|
|
1339
|
-
|
|
1340
|
-
# track xml:base and xml:lang
|
|
1341
|
-
baseuri = attrsD['xml:base'] || attrsD['base'] || @baseuri
|
|
1342
|
-
@baseuri = urljoin(@baseuri, baseuri)
|
|
1343
|
-
lang = attrsD['xml:lang'] || attrsD['lang']
|
|
1344
|
-
if lang == '' # FIXME This next bit of code is right? Wtf?
|
|
1345
|
-
# xml:lang could be explicitly set to '', we need to capture that
|
|
1346
|
-
lang = nil
|
|
1347
|
-
elsif lang.nil?
|
|
1348
|
-
# if no xml:lang is specified, use parent lang
|
|
1349
|
-
lang = @lang
|
|
1350
|
-
end
|
|
1351
|
-
if lang and not lang.empty? # Seriously, this cannot be correct
|
|
1352
|
-
if ['feed', 'rss', 'rdf:RDF'].include?tag
|
|
1353
|
-
@feeddata['language'] = lang.gsub('_','-')
|
|
1354
|
-
end
|
|
1355
|
-
end
|
|
1356
|
-
@lang = lang
|
|
1357
|
-
@basestack << @baseuri
|
|
1358
|
-
@langstack << lang
|
|
1359
|
-
|
|
1360
|
-
# track namespaces
|
|
1361
|
-
attrsd.each do |prefix, uri|
|
|
1362
|
-
if /^xmlns:/ =~ prefix # prefix begins with xmlns:
|
|
1363
|
-
trackNamespace(prefix[6..-1], uri)
|
|
1364
|
-
elsif prefix == 'xmlns':
|
|
1365
|
-
trackNamespace(nil, uri)
|
|
1366
|
-
end
|
|
1367
|
-
end
|
|
1368
|
-
|
|
1369
|
-
# track inline content
|
|
1370
|
-
if @incontent != 0 and @contentparams.has_key?('type') and not ( /xml$/ =~ (@contentparams['type'] || 'xml') )
|
|
1371
|
-
# element declared itself as escaped markup, but isn't really
|
|
1372
|
-
|
|
1373
|
-
@contentparams['type'] = 'application/xhtml+xml'
|
|
1374
|
-
end
|
|
1375
|
-
if @incontent != 0 and @contentparams['type'] == 'application/xhtml+xml'
|
|
1376
|
-
# Note: probably shouldn't simply recreate localname here, but
|
|
1377
|
-
# our namespace handling isn't actually 100% correct in cases where
|
|
1378
|
-
# the feed redefines the default namespace (which is actually
|
|
1379
|
-
# the usual case for inline content, thanks Sam), so here we
|
|
1380
|
-
# cheat and just reconstruct the element based on localname
|
|
1381
|
-
# because that compensates for the bugs in our namespace handling.
|
|
1382
|
-
# This will horribly munge inline content with non-empty qnames,
|
|
1383
|
-
# but nobody actually does that, so I'm not fixing it.
|
|
1384
|
-
tag = tag.split(':')[-1]
|
|
1385
|
-
attrsA = attrsd.to_a.collect{|l| "#{l[0]}=\"#{l[1]}\""}
|
|
1386
|
-
attrsS = ' '+attrsA.join(' ')
|
|
1387
|
-
return handle_data("<#{tag}#{attrsS}>", escape=false)
|
|
1388
|
-
end
|
|
1389
|
-
|
|
1390
|
-
# match namespaces
|
|
1391
|
-
if /:/ =~ tag
|
|
1392
|
-
prefix, suffix = tag.split(':', 2)
|
|
1393
|
-
else
|
|
1394
|
-
prefix, suffix = '', tag
|
|
1395
|
-
end
|
|
1396
|
-
prefix = @namespacemap[prefix] || prefix
|
|
1397
|
-
if prefix and not prefix.empty?
|
|
1398
|
-
prefix = prefix + '_'
|
|
1399
|
-
end
|
|
1400
|
-
|
|
1401
|
-
# special hack for better tracking of empty textinput/image elements in illformed feeds
|
|
1402
|
-
if (not prefix and not prefix.empty?) and not (['title', 'link', 'description','name'].include?tag)
|
|
1403
|
-
@intextinput = false
|
|
1404
|
-
end
|
|
1405
|
-
if (prefix.nil? or prefix.empty?) and not (['title', 'link', 'description', 'url', 'href', 'width', 'height'].include?tag)
|
|
1406
|
-
@inimage = false
|
|
1407
|
-
end
|
|
1408
|
-
|
|
1409
|
-
# call special handler (if defined) or default handler
|
|
1410
|
-
begin
|
|
1411
|
-
return send('_start_'+prefix+suffix, attrsD)
|
|
1412
|
-
rescue NoMethodError
|
|
1413
|
-
return push(prefix + suffix, true)
|
|
1414
|
-
end
|
|
1415
|
-
end # End unknown_starttag
|
|
1416
|
-
|
|
1417
|
-
def unknown_endtag(tag)
|
|
1418
|
-
$stderr << "end #{tag}\n" if $debug
|
|
1419
|
-
# match namespaces
|
|
1420
|
-
if tag.index(':')
|
|
1421
|
-
prefix, suffix = tag.split(':',2)
|
|
1422
|
-
else
|
|
1423
|
-
prefix, suffix = '', tag
|
|
1424
|
-
end
|
|
1425
|
-
prefix = @namespacemap[prefix] || prefix
|
|
1426
|
-
if prefix and not prefix.empty?
|
|
1427
|
-
prefix = prefix + '_'
|
|
1428
|
-
end
|
|
1429
|
-
|
|
1430
|
-
# call special handler (if defined) or default handler
|
|
1431
|
-
begin
|
|
1432
|
-
send('_end_' + prefix + suffix) # NOTE no return here! do not add it!
|
|
1433
|
-
rescue NoMethodError => details
|
|
1434
|
-
pop(prefix + suffix)
|
|
1435
|
-
end
|
|
1436
|
-
|
|
1437
|
-
# track inline content
|
|
1438
|
-
if @incontent != 0 and @contentparams.has_key?'type' and /xml$/ =~ (@contentparams['type'] || 'xml')
|
|
1439
|
-
# element declared itself as escaped markup, but it isn't really
|
|
1440
|
-
@contentparams['type'] = 'application/xhtml+xml'
|
|
1441
|
-
end
|
|
1442
|
-
if @incontent != 0 and @contentparams['type'] == 'application/xhtml+xml'
|
|
1443
|
-
tag = tag.split(':')[-1]
|
|
1444
|
-
handle_data("</#{tag}>", escape=false)
|
|
1445
|
-
end
|
|
1446
|
-
|
|
1447
|
-
# track xml:base and xml:lang going out of scope
|
|
1448
|
-
if @basestack and not @basestack.empty?
|
|
1449
|
-
@basestack.pop
|
|
1450
|
-
if @basestack and @basestack[-1] and not (@basestack.empty? or @basestack[-1].empty?)
|
|
1451
|
-
@baseuri = @basestack[-1]
|
|
1452
|
-
end
|
|
1453
|
-
end
|
|
1454
|
-
if @langstack and not @langstack.empty?
|
|
1455
|
-
@langstack.pop
|
|
1456
|
-
if @langstack and not @langstack.empty? # and @langstack[-1] and not @langstack.empty?
|
|
1457
|
-
@lang = @langstack[-1]
|
|
1458
|
-
end
|
|
1459
|
-
end
|
|
1460
|
-
end
|
|
1461
|
-
|
|
1462
|
-
def handle_charref(ref)
|
|
1463
|
-
# LooseParserOnly
|
|
1464
|
-
# called for each character reference, e.g. for ' ', ref will be '160'
|
|
1465
|
-
$stderr << "entering handle_charref with #{ref}\n" if $debug
|
|
1466
|
-
return if @elementstack.nil? or @elementstack.empty?
|
|
1467
|
-
ref.downcase!
|
|
1468
|
-
chars = ['34', '38', '39', '60', '62', 'x22', 'x26', 'x27', 'x3c', 'x3e']
|
|
1469
|
-
if chars.include?ref
|
|
1470
|
-
text = "&##{ref};"
|
|
1471
|
-
else
|
|
1472
|
-
if ref[0..0] == 'x'
|
|
1473
|
-
c = (ref[1..-1]).to_i(16)
|
|
1474
|
-
else
|
|
1475
|
-
c = ref.to_i
|
|
1476
|
-
end
|
|
1477
|
-
text = uconvert(unichr(c),'unicode')
|
|
1478
|
-
end
|
|
1479
|
-
@elementstack[-1][2] << text
|
|
1480
|
-
end
|
|
1481
|
-
|
|
1482
|
-
def handle_entityref(ref)
|
|
1483
|
-
# LooseParserOnly
|
|
1484
|
-
# called for each entity reference, e.g. for '©', ref will be 'copy'
|
|
1485
|
-
|
|
1486
|
-
return if @elementstack.nil? or @elementstack.empty?
|
|
1487
|
-
$stderr << "entering handle_entityref with #{ref}\n" if $debug
|
|
1488
|
-
ents = ['lt', 'gt', 'quot', 'amp', 'apos']
|
|
1489
|
-
if ents.include?ref
|
|
1490
|
-
text = "&#{ref};"
|
|
1491
|
-
else
|
|
1492
|
-
text = HTMLEntities::decode_entities("&#{ref};")
|
|
1493
|
-
end
|
|
1494
|
-
@elementstack[-1][2] << text
|
|
1495
|
-
end
|
|
1496
|
-
|
|
1497
|
-
def handle_data(text, escape=true)
|
|
1498
|
-
# called for each block of plain text, i.e. outside of any tag and
|
|
1499
|
-
# not containing any character or entity references
|
|
1500
|
-
return if @elementstack.nil? or @elementstack.empty?
|
|
1501
|
-
if escape and @contentparams['type'] == 'application/xhtml+xml'
|
|
1502
|
-
text = text.to_xs
|
|
1503
|
-
end
|
|
1504
|
-
@elementstack[-1][2] << text
|
|
1505
|
-
end
|
|
1506
|
-
|
|
1507
|
-
def handle_comment(comment)
|
|
1508
|
-
# called for each comment, e.g. <!-- insert message here -->
|
|
1509
|
-
end
|
|
1510
|
-
|
|
1511
|
-
def handle_pi(text)
|
|
1512
|
-
end
|
|
1513
|
-
|
|
1514
|
-
def handle_decl(text)
|
|
1515
|
-
end
|
|
1516
|
-
|
|
1517
|
-
def parse_declaration(i)
|
|
1518
|
-
# for LooseFeedParser
|
|
1519
|
-
$stderr << "entering parse_declaration\n" if $debug
|
|
1520
|
-
if @rawdata[i...i+9] == '<![CDATA['
|
|
1521
|
-
k = @rawdata.index(/\]\]>/u,i+9)
|
|
1522
|
-
k = @rawdata.length unless k
|
|
1523
|
-
handle_data(@rawdata[i+9...k].to_xs,false)
|
|
1524
|
-
return k+3
|
|
1525
|
-
else
|
|
1526
|
-
k = @rawdata.index(/>/,i).to_i
|
|
1527
|
-
return k+1
|
|
1528
|
-
end
|
|
1529
|
-
end
|
|
1530
|
-
|
|
1531
|
-
def mapContentType(contentType)
|
|
1532
|
-
contentType.downcase!
|
|
1533
|
-
case contentType
|
|
1534
|
-
when 'text'
|
|
1535
|
-
contentType = 'text/plain'
|
|
1536
|
-
when 'html'
|
|
1537
|
-
contentType = 'text/html'
|
|
1538
|
-
when 'xhtml'
|
|
1539
|
-
contentType = 'application/xhtml+xml'
|
|
1540
|
-
end
|
|
1541
|
-
return contentType
|
|
1542
|
-
end
|
|
1543
|
-
|
|
1544
|
-
def trackNamespace(prefix, uri)
|
|
1545
|
-
|
|
1546
|
-
loweruri = uri.downcase.strip
|
|
1547
|
-
if [prefix, loweruri] == [nil, 'http://my.netscape.com/rdf/simple/0.9/'] and (@version.nil? or @version.empty?)
|
|
1548
|
-
@version = 'rss090'
|
|
1549
|
-
elsif loweruri == 'http://purl.org/rss/1.0/' and (@version.nil? or @version.empty?)
|
|
1550
|
-
@version = 'rss10'
|
|
1551
|
-
elsif loweruri == 'http://www.w3.org/2005/atom' and (@version.nil? or @version.empty?)
|
|
1552
|
-
@version = 'atom10'
|
|
1553
|
-
elsif /backend\.userland\.com\/rss/ =~ loweruri
|
|
1554
|
-
# match any backend.userland.com namespace
|
|
1555
|
-
uri = 'http://backend.userland.com/rss'
|
|
1556
|
-
loweruri = uri
|
|
1557
|
-
end
|
|
1558
|
-
if @matchnamespaces.has_key? loweruri
|
|
1559
|
-
@namespacemap[prefix] = @matchnamespaces[loweruri]
|
|
1560
|
-
@namespacesInUse[@matchnamespaces[loweruri]] = uri
|
|
1561
|
-
else
|
|
1562
|
-
@namespacesInUse[prefix || ''] = uri
|
|
1563
|
-
end
|
|
1564
|
-
end
|
|
1565
|
-
|
|
1566
|
-
def resolveURI(uri)
|
|
1567
|
-
return urljoin(@baseuri || '', uri)
|
|
1568
|
-
end
|
|
1569
|
-
|
|
1570
|
-
def decodeEntities(element, data)
|
|
1571
|
-
return data
|
|
1572
|
-
end
|
|
1573
|
-
|
|
1574
|
-
def push(element, expectingText)
|
|
1575
|
-
@elementstack << [element, expectingText, []]
|
|
1576
|
-
end
|
|
1577
|
-
|
|
1578
|
-
def pop(element, stripWhitespace=true)
|
|
1579
|
-
return if @elementstack.nil? or @elementstack.empty?
|
|
1580
|
-
return if @elementstack[-1][0] != element
|
|
1581
|
-
element, expectingText, pieces = @elementstack.pop
|
|
1582
|
-
if pieces.class == Array
|
|
1583
|
-
output = pieces.join('')
|
|
1584
|
-
else
|
|
1585
|
-
output = pieces
|
|
1586
|
-
end
|
|
1587
|
-
if stripWhitespace
|
|
1588
|
-
output.strip!
|
|
1589
|
-
end
|
|
1590
|
-
return output if not expectingText
|
|
1591
|
-
|
|
1592
|
-
# decode base64 content
|
|
1593
|
-
if @contentparams['base64']
|
|
1594
|
-
out64 = Base64::decode64(output) # a.k.a. [output].unpack('m')[0]
|
|
1595
|
-
if not output.empty? and not out64.empty?
|
|
1596
|
-
output = out64
|
|
1597
|
-
end
|
|
1598
|
-
end
|
|
1599
|
-
|
|
1600
|
-
# resolve relative URIs
|
|
1601
|
-
if @can_be_relative_uri.include?element and output and not output.empty?
|
|
1602
|
-
output = resolveURI(output)
|
|
1603
|
-
end
|
|
1604
|
-
|
|
1605
|
-
# decode entities within embedded markup
|
|
1606
|
-
if not @contentparams['base64']
|
|
1607
|
-
output = decodeEntities(element, output)
|
|
1608
|
-
end
|
|
1609
|
-
|
|
1610
|
-
# remove temporary cruft from contentparams
|
|
1611
|
-
@contentparams.delete('mode')
|
|
1612
|
-
@contentparams.delete('base64')
|
|
1613
|
-
|
|
1614
|
-
# resolve relative URIs within embedded markup
|
|
1615
|
-
if @html_types.include?mapContentType(@contentparams['type'] || 'text/html')
|
|
1616
|
-
if @can_contain_relative_uris.include?element
|
|
1617
|
-
output = FeedParser.resolveRelativeURIs(output, @baseuri, @encoding)
|
|
1618
|
-
end
|
|
1619
|
-
end
|
|
1620
|
-
# sanitize embedded markup
|
|
1621
|
-
if @html_types.include?mapContentType(@contentparams['type'] || 'text/html')
|
|
1622
|
-
if @can_contain_dangerous_markup.include?element
|
|
1623
|
-
output = FeedParser.sanitizeHTML(output, @encoding)
|
|
1624
|
-
end
|
|
1625
|
-
end
|
|
1626
|
-
|
|
1627
|
-
if @encoding and not @encoding.empty? and @encoding != 'utf-8'
|
|
1628
|
-
output = uconvert(output, @encoding, 'utf-8')
|
|
1629
|
-
# FIXME I turn everything into utf-8, not unicode, originally because REXML was being used but now beause I haven't tested it out yet.
|
|
1630
|
-
end
|
|
1631
|
-
|
|
1632
|
-
# categories/tags/keywords/whatever are handled in _end_category
|
|
1633
|
-
return output if element == 'category'
|
|
1634
|
-
|
|
1635
|
-
# store output in appropriate place(s)
|
|
1636
|
-
if @inentry and not @insource
|
|
1637
|
-
if element == 'content'
|
|
1638
|
-
@entries[-1][element] ||= []
|
|
1639
|
-
contentparams = Marshal.load(Marshal.dump(@contentparams)) # deepcopy
|
|
1640
|
-
contentparams['value'] = output
|
|
1641
|
-
@entries[-1][element] << contentparams
|
|
1642
|
-
elsif element == 'link'
|
|
1643
|
-
@entries[-1][element] = output
|
|
1644
|
-
if output and not output.empty?
|
|
1645
|
-
@entries[-1]['links'][-1]['href'] = output
|
|
1646
|
-
end
|
|
1647
|
-
else
|
|
1648
|
-
element = 'summary' if element == 'description'
|
|
1649
|
-
@entries[-1][element] = output
|
|
1650
|
-
if @incontent != 0
|
|
1651
|
-
contentparams = Marshal.load(Marshal.dump(@contentparams))
|
|
1652
|
-
contentparams['value'] = output
|
|
1653
|
-
@entries[-1][element + '_detail'] = contentparams
|
|
1654
|
-
end
|
|
1655
|
-
end
|
|
1656
|
-
elsif (@infeed or @insource) and not @intextinput and not @inimage
|
|
1657
|
-
context = getContext()
|
|
1658
|
-
element = 'subtitle' if element == 'description'
|
|
1659
|
-
context[element] = output
|
|
1660
|
-
if element == 'link'
|
|
1661
|
-
context['links'][-1]['href'] = output
|
|
1662
|
-
elsif @incontent != 0
|
|
1663
|
-
contentparams = Marshal.load(Marshal.dump(@contentparams))
|
|
1664
|
-
contentparams['value'] = output
|
|
1665
|
-
context[element + '_detail'] = contentparams
|
|
1666
|
-
end
|
|
1667
|
-
end
|
|
1668
|
-
return output
|
|
1669
|
-
end
|
|
1670
|
-
|
|
1671
|
-
def pushContent(tag, attrsD, defaultContentType, expectingText)
|
|
1672
|
-
@incontent += 1 # Yes, I hate this.
|
|
1673
|
-
type = mapContentType(attrsD['type'] || defaultContentType)
|
|
1674
|
-
@contentparams = FeedParserDict.new({'type' => type,'language' => @lang,'base' => @baseuri})
|
|
1675
|
-
@contentparams['base64'] = isBase64(attrsD, @contentparams)
|
|
1676
|
-
push(tag, expectingText)
|
|
1677
|
-
end
|
|
1678
|
-
|
|
1679
|
-
def popContent(tag)
|
|
1680
|
-
value = pop(tag)
|
|
1681
|
-
@incontent -= 1
|
|
1682
|
-
@contentparams.clear
|
|
1683
|
-
return value
|
|
1684
|
-
end
|
|
1685
|
-
|
|
1686
|
-
def mapToStandardPrefix(name)
|
|
1687
|
-
colonpos = name.index(':')
|
|
1688
|
-
if colonpos
|
|
1689
|
-
prefix = name[0..colonpos-1]
|
|
1690
|
-
suffix = name[colonpos+1..-1]
|
|
1691
|
-
prefix = @namespacemap[prefix] || prefix
|
|
1692
|
-
name = prefix + ':' + suffix
|
|
1693
|
-
end
|
|
1694
|
-
return name
|
|
1695
|
-
end
|
|
1696
|
-
|
|
1697
|
-
def getAttribute(attrsD, name)
|
|
1698
|
-
return attrsD[mapToStandardPrefix(name)]
|
|
1699
|
-
end
|
|
1700
|
-
|
|
1701
|
-
def isBase64(attrsD, contentparams)
|
|
1702
|
-
return true if (attrsD['mode'] == 'base64')
|
|
1703
|
-
if /(^text\/)|(\+xml$)|(\/xml$)/ =~ contentparams['type']
|
|
1704
|
-
return false
|
|
1705
|
-
end
|
|
1706
|
-
return true
|
|
1707
|
-
end
|
|
1708
|
-
|
|
1709
|
-
def itsAnHrefDamnIt(attrsD)
|
|
1710
|
-
href= attrsD['url'] || attrsD['uri'] || attrsD['href']
|
|
1711
|
-
if href
|
|
1712
|
-
attrsD.delete('url')
|
|
1713
|
-
attrsD.delete('uri')
|
|
1714
|
-
attrsD['href'] = href
|
|
1715
|
-
end
|
|
1716
|
-
return attrsD
|
|
1717
|
-
end
|
|
1718
|
-
|
|
1719
|
-
|
|
1720
|
-
def _save(key, value)
|
|
1721
|
-
context = getContext()
|
|
1722
|
-
context[key] ||= value
|
|
1723
|
-
end
|
|
1724
|
-
|
|
1725
|
-
def _start_rss(attrsD)
|
|
1726
|
-
versionmap = {'0.91' => 'rss091u',
|
|
1727
|
-
'0.92' => 'rss092',
|
|
1728
|
-
'0.93' => 'rss093',
|
|
1729
|
-
'0.94' => 'rss094'
|
|
1730
|
-
}
|
|
1731
|
-
|
|
1732
|
-
if not @version or @version.empty?
|
|
1733
|
-
attr_version = attrsD['version'] || ''
|
|
1734
|
-
version = versionmap[attr_version]
|
|
1735
|
-
if version and not version.empty?
|
|
1736
|
-
@version = version
|
|
1737
|
-
elsif /^2\./ =~ attr_version
|
|
1738
|
-
@version = 'rss20'
|
|
1739
|
-
else
|
|
1740
|
-
@version = 'rss'
|
|
1741
|
-
end
|
|
1742
|
-
end
|
|
1743
|
-
end
|
|
1744
|
-
|
|
1745
|
-
def _start_dlhottitles(attrsD)
|
|
1746
|
-
@version = 'hotrss'
|
|
1747
|
-
end
|
|
1748
|
-
|
|
1749
|
-
def _start_channel(attrsD)
|
|
1750
|
-
@infeed = true
|
|
1751
|
-
_cdf_common(attrsD)
|
|
1752
|
-
end
|
|
1753
|
-
alias :_start_feedinfo :_start_channel
|
|
1754
|
-
|
|
1755
|
-
def _cdf_common(attrsD)
|
|
1756
|
-
if attrsD.has_key?'lastmod'
|
|
1757
|
-
_start_modified({})
|
|
1758
|
-
@elementstack[-1][-1] = attrsD['lastmod']
|
|
1759
|
-
_end_modified
|
|
1760
|
-
end
|
|
1761
|
-
if attrsD.has_key?'href'
|
|
1762
|
-
_start_link({})
|
|
1763
|
-
@elementstack[-1][-1] = attrsD['href']
|
|
1764
|
-
_end_link
|
|
1765
|
-
end
|
|
1766
|
-
end
|
|
1767
|
-
|
|
1768
|
-
def _start_feed(attrsD)
|
|
1769
|
-
@infeed = true
|
|
1770
|
-
versionmap = {'0.1' => 'atom01',
|
|
1771
|
-
'0.2' => 'atom02',
|
|
1772
|
-
'0.3' => 'atom03'
|
|
1773
|
-
}
|
|
1774
|
-
|
|
1775
|
-
if not @version or @version.empty?
|
|
1776
|
-
attr_version = attrsD['version']
|
|
1777
|
-
version = versionmap[attr_version]
|
|
1778
|
-
if @version and not @version.empty?
|
|
1779
|
-
@version = version
|
|
1780
|
-
else
|
|
1781
|
-
@version = 'atom'
|
|
1782
|
-
end
|
|
1783
|
-
end
|
|
1784
|
-
end
|
|
1785
|
-
|
|
1786
|
-
def _end_channel
|
|
1787
|
-
@infeed = false
|
|
1788
|
-
end
|
|
1789
|
-
alias :_end_feed :_end_channel
|
|
1790
|
-
|
|
1791
|
-
def _start_image(attrsD)
|
|
1792
|
-
@inimage = true
|
|
1793
|
-
push('image', false)
|
|
1794
|
-
context = getContext()
|
|
1795
|
-
context['image'] ||= FeedParserDict.new
|
|
1796
|
-
end
|
|
1797
|
-
|
|
1798
|
-
def _end_image
|
|
1799
|
-
pop('image')
|
|
1800
|
-
@inimage = false
|
|
1801
|
-
end
|
|
1802
|
-
|
|
1803
|
-
def _start_textinput(attrsD)
|
|
1804
|
-
@intextinput = true
|
|
1805
|
-
push('textinput', false)
|
|
1806
|
-
context = getContext()
|
|
1807
|
-
context['textinput'] ||= FeedParserDict.new
|
|
1808
|
-
end
|
|
1809
|
-
alias :_start_textInput :_start_textinput
|
|
1810
|
-
|
|
1811
|
-
def _end_textinput
|
|
1812
|
-
pop('textinput')
|
|
1813
|
-
@intextinput = false
|
|
1814
|
-
end
|
|
1815
|
-
alias :_end_textInput :_end_textinput
|
|
1816
|
-
|
|
1817
|
-
def _start_author(attrsD)
|
|
1818
|
-
@inauthor = true
|
|
1819
|
-
push('author', true)
|
|
1820
|
-
end
|
|
1821
|
-
alias :_start_managingeditor :_start_author
|
|
1822
|
-
alias :_start_dc_author :_start_author
|
|
1823
|
-
alias :_start_dc_creator :_start_author
|
|
1824
|
-
alias :_start_itunes_author :_start_author
|
|
1825
|
-
|
|
1826
|
-
def _end_author
|
|
1827
|
-
pop('author')
|
|
1828
|
-
@inauthor = false
|
|
1829
|
-
_sync_author_detail()
|
|
1830
|
-
end
|
|
1831
|
-
alias :_end_managingeditor :_end_author
|
|
1832
|
-
alias :_end_dc_author :_end_author
|
|
1833
|
-
alias :_end_dc_creator :_end_author
|
|
1834
|
-
alias :_end_itunes_author :_end_author
|
|
1835
|
-
|
|
1836
|
-
def _start_itunes_owner(attrsD)
|
|
1837
|
-
@inpublisher = true
|
|
1838
|
-
push('publisher', false)
|
|
1839
|
-
end
|
|
1840
|
-
|
|
1841
|
-
def _end_itunes_owner
|
|
1842
|
-
pop('publisher')
|
|
1843
|
-
@inpublisher = false
|
|
1844
|
-
_sync_author_detail('publisher')
|
|
1845
|
-
end
|
|
1846
|
-
|
|
1847
|
-
def _start_contributor(attrsD)
|
|
1848
|
-
@incontributor = true
|
|
1849
|
-
context = getContext()
|
|
1850
|
-
context['contributors'] ||= []
|
|
1851
|
-
context['contributors'] << FeedParserDict.new
|
|
1852
|
-
push('contributor', false)
|
|
1853
|
-
end
|
|
1854
|
-
|
|
1855
|
-
def _end_contributor
|
|
1856
|
-
pop('contributor')
|
|
1857
|
-
@incontributor = false
|
|
1858
|
-
end
|
|
1859
|
-
|
|
1860
|
-
def _start_dc_contributor(attrsD)
|
|
1861
|
-
@incontributor = true
|
|
1862
|
-
context = getContext()
|
|
1863
|
-
context['contributors'] ||= []
|
|
1864
|
-
context['contributors'] << FeedParserDict.new
|
|
1865
|
-
push('name', false)
|
|
1866
|
-
end
|
|
1867
|
-
|
|
1868
|
-
def _end_dc_contributor
|
|
1869
|
-
_end_name
|
|
1870
|
-
@incontributor = false
|
|
1871
|
-
end
|
|
1872
|
-
|
|
1873
|
-
def _start_name(attrsD)
|
|
1874
|
-
push('name', false)
|
|
1875
|
-
end
|
|
1876
|
-
alias :_start_itunes_name :_start_name
|
|
1877
|
-
|
|
1878
|
-
def _end_name
|
|
1879
|
-
value = pop('name')
|
|
1880
|
-
if @inpublisher
|
|
1881
|
-
_save_author('name', value, 'publisher')
|
|
1882
|
-
elsif @inauthor
|
|
1883
|
-
_save_author('name', value)
|
|
1884
|
-
elsif @incontributor
|
|
1885
|
-
_save_contributor('name', value)
|
|
1886
|
-
elsif @intextinput
|
|
1887
|
-
context = getContext()
|
|
1888
|
-
context['textinput']['name'] = value
|
|
1889
|
-
end
|
|
1890
|
-
end
|
|
1891
|
-
alias :_end_itunes_name :_end_name
|
|
1892
|
-
|
|
1893
|
-
def _start_width(attrsD)
|
|
1894
|
-
push('width', false)
|
|
1895
|
-
end
|
|
1896
|
-
|
|
1897
|
-
def _end_width
|
|
1898
|
-
value = pop('width').to_i
|
|
1899
|
-
if @inimage
|
|
1900
|
-
context = getContext
|
|
1901
|
-
context['image']['width'] = value
|
|
1902
|
-
end
|
|
1903
|
-
end
|
|
1904
|
-
|
|
1905
|
-
def _start_height(attrsD)
|
|
1906
|
-
push('height', false)
|
|
1907
|
-
end
|
|
1908
|
-
|
|
1909
|
-
def _end_height
|
|
1910
|
-
value = pop('height').to_i
|
|
1911
|
-
if @inimage
|
|
1912
|
-
context = getContext()
|
|
1913
|
-
context['image']['height'] = value
|
|
1914
|
-
end
|
|
1915
|
-
end
|
|
1916
|
-
|
|
1917
|
-
def _start_url(attrsD)
|
|
1918
|
-
push('href', true)
|
|
1919
|
-
end
|
|
1920
|
-
alias :_start_homepage :_start_url
|
|
1921
|
-
alias :_start_uri :_start_url
|
|
1922
|
-
|
|
1923
|
-
def _end_url
|
|
1924
|
-
value = pop('href')
|
|
1925
|
-
if @inauthor
|
|
1926
|
-
_save_author('href', value)
|
|
1927
|
-
elsif @incontributor
|
|
1928
|
-
_save_contributor('href', value)
|
|
1929
|
-
elsif @inimage
|
|
1930
|
-
context = getContext()
|
|
1931
|
-
context['image']['href'] = value
|
|
1932
|
-
elsif @intextinput
|
|
1933
|
-
context = getContext()
|
|
1934
|
-
context['textinput']['link'] = value
|
|
1935
|
-
end
|
|
1936
|
-
end
|
|
1937
|
-
alias :_end_homepage :_end_url
|
|
1938
|
-
alias :_end_uri :_end_url
|
|
1939
|
-
|
|
1940
|
-
def _start_email(attrsD)
|
|
1941
|
-
push('email', false)
|
|
1942
|
-
end
|
|
1943
|
-
alias :_start_itunes_email :_start_email
|
|
1944
|
-
|
|
1945
|
-
def _end_email
|
|
1946
|
-
value = pop('email')
|
|
1947
|
-
if @inpublisher
|
|
1948
|
-
_save_author('email', value, 'publisher')
|
|
1949
|
-
elsif @inauthor
|
|
1950
|
-
_save_author('email', value)
|
|
1951
|
-
elsif @incontributor
|
|
1952
|
-
_save_contributor('email', value)
|
|
1953
|
-
end
|
|
1954
|
-
end
|
|
1955
|
-
alias :_end_itunes_email :_end_email
|
|
1956
|
-
|
|
1957
|
-
def getContext
|
|
1958
|
-
if @insource
|
|
1959
|
-
context = @sourcedata
|
|
1960
|
-
elsif @inentry
|
|
1961
|
-
context = @entries[-1]
|
|
1962
|
-
else
|
|
1963
|
-
context = @feeddata
|
|
1964
|
-
end
|
|
1965
|
-
return context
|
|
1966
|
-
end
|
|
1967
|
-
|
|
1968
|
-
def _save_author(key, value, prefix='author')
|
|
1969
|
-
context = getContext()
|
|
1970
|
-
context[prefix + '_detail'] ||= FeedParserDict.new
|
|
1971
|
-
context[prefix + '_detail'][key] = value
|
|
1972
|
-
_sync_author_detail()
|
|
1973
|
-
end
|
|
1974
|
-
|
|
1975
|
-
def _save_contributor(key, value)
|
|
1976
|
-
context = getContext
|
|
1977
|
-
context['contributors'] ||= [FeedParserDict.new]
|
|
1978
|
-
context['contributors'][-1][key] = value
|
|
1979
|
-
end
|
|
1980
|
-
|
|
1981
|
-
def _sync_author_detail(key='author')
|
|
1982
|
-
context = getContext()
|
|
1983
|
-
detail = context["#{key}_detail"]
|
|
1984
|
-
if detail and not detail.empty?
|
|
1985
|
-
name = detail['name']
|
|
1986
|
-
email = detail['email']
|
|
1987
|
-
|
|
1988
|
-
if name and email and not (name.empty? or name.empty?)
|
|
1989
|
-
context[key] = "#{name} (#{email})"
|
|
1990
|
-
elsif name and not name.empty?
|
|
1991
|
-
context[key] = name
|
|
1992
|
-
elsif email and not email.empty?
|
|
1993
|
-
context[key] = email
|
|
1994
|
-
end
|
|
1995
|
-
else
|
|
1996
|
-
author = context[key].dup unless context[key].nil?
|
|
1997
|
-
return if not author or author.empty?
|
|
1998
|
-
emailmatch = author.match(/(([a-zA-Z0-9\_\-\.\+]+)@((\[[0-9]{1,3}\.[0-9]{1,3}\.[0-9]{1,3}\.)|(([a-zA-Z0-9\-]+\.)+))([a-zA-Z]{2,4}|[0-9]{1,3})(\]?))/)
|
|
1999
|
-
email = emailmatch[1]
|
|
2000
|
-
author.gsub!(email, '')
|
|
2001
|
-
author.gsub!("\(\)", '')
|
|
2002
|
-
author.strip!
|
|
2003
|
-
author.gsub!(/^\(/,'')
|
|
2004
|
-
author.gsub!(/\)$/,'')
|
|
2005
|
-
author.strip!
|
|
2006
|
-
context["#{key}_detail"] ||= FeedParserDict.new
|
|
2007
|
-
context["#{key}_detail"]['name'] = author
|
|
2008
|
-
context["#{key}_detail"]['email'] = email
|
|
2009
|
-
end
|
|
2010
|
-
end
|
|
2011
|
-
|
|
2012
|
-
def _start_subtitle(attrsD)
|
|
2013
|
-
pushContent('subtitle', attrsD, 'text/plain', true)
|
|
2014
|
-
end
|
|
2015
|
-
alias :_start_tagline :_start_subtitle
|
|
2016
|
-
alias :_start_itunes_subtitle :_start_subtitle
|
|
2017
|
-
|
|
2018
|
-
def _end_subtitle
|
|
2019
|
-
popContent('subtitle')
|
|
2020
|
-
end
|
|
2021
|
-
alias :_end_tagline :_end_subtitle
|
|
2022
|
-
alias :_end_itunes_subtitle :_end_subtitle
|
|
2023
|
-
|
|
2024
|
-
def _start_rights(attrsD)
|
|
2025
|
-
pushContent('rights', attrsD, 'text/plain', true)
|
|
2026
|
-
end
|
|
2027
|
-
alias :_start_dc_rights :_start_rights
|
|
2028
|
-
alias :_start_copyright :_start_rights
|
|
2029
|
-
|
|
2030
|
-
def _end_rights
|
|
2031
|
-
popContent('rights')
|
|
2032
|
-
end
|
|
2033
|
-
alias :_end_dc_rights :_end_rights
|
|
2034
|
-
alias :_end_copyright :_end_rights
|
|
2035
|
-
|
|
2036
|
-
def _start_item(attrsD)
|
|
2037
|
-
@entries << FeedParserDict.new
|
|
2038
|
-
push('item', false)
|
|
2039
|
-
@inentry = true
|
|
2040
|
-
@guidislink = false
|
|
2041
|
-
id = getAttribute(attrsD, 'rdf:about')
|
|
2042
|
-
if id and not id.empty?
|
|
2043
|
-
context = getContext()
|
|
2044
|
-
context['id'] = id
|
|
2045
|
-
end
|
|
2046
|
-
_cdf_common(attrsD)
|
|
2047
|
-
end
|
|
2048
|
-
alias :_start_entry :_start_item
|
|
2049
|
-
alias :_start_product :_start_item
|
|
2050
|
-
|
|
2051
|
-
def _end_item
|
|
2052
|
-
pop('item')
|
|
2053
|
-
@inentry = false
|
|
2054
|
-
end
|
|
2055
|
-
alias :_end_entry :_end_item
|
|
2056
|
-
|
|
2057
|
-
def _start_dc_language(attrsD)
|
|
2058
|
-
push('language', true)
|
|
2059
|
-
end
|
|
2060
|
-
alias :_start_language :_start_dc_language
|
|
2061
|
-
|
|
2062
|
-
def _end_dc_language
|
|
2063
|
-
@lang = pop('language')
|
|
2064
|
-
end
|
|
2065
|
-
alias :_end_language :_end_dc_language
|
|
2066
|
-
|
|
2067
|
-
def _start_dc_publisher(attrsD)
|
|
2068
|
-
push('publisher', true)
|
|
2069
|
-
end
|
|
2070
|
-
alias :_start_webmaster :_start_dc_publisher
|
|
2071
|
-
|
|
2072
|
-
def _end_dc_publisher
|
|
2073
|
-
pop('publisher')
|
|
2074
|
-
_sync_author_detail('publisher')
|
|
2075
|
-
end
|
|
2076
|
-
alias :_end_webmaster :_end_dc_publisher
|
|
2077
|
-
|
|
2078
|
-
def _start_published(attrsD)
|
|
2079
|
-
push('published', true)
|
|
2080
|
-
end
|
|
2081
|
-
alias :_start_dcterms_issued :_start_published
|
|
2082
|
-
alias :_start_issued :_start_published
|
|
2083
|
-
|
|
2084
|
-
def _end_published
|
|
2085
|
-
value = pop('published')
|
|
2086
|
-
_save('published_parsed', parse_date(value))
|
|
2087
|
-
end
|
|
2088
|
-
alias :_end_dcterms_issued :_end_published
|
|
2089
|
-
alias :_end_issued :_end_published
|
|
2090
|
-
|
|
2091
|
-
def _start_updated(attrsD)
|
|
2092
|
-
push('updated', true)
|
|
2093
|
-
end
|
|
2094
|
-
alias :_start_modified :_start_updated
|
|
2095
|
-
alias :_start_dcterms_modified :_start_updated
|
|
2096
|
-
alias :_start_pubdate :_start_updated
|
|
2097
|
-
alias :_start_dc_date :_start_updated
|
|
2098
|
-
|
|
2099
|
-
def _end_updated
|
|
2100
|
-
value = pop('updated')
|
|
2101
|
-
_save('updated_parsed', parse_date(value))
|
|
2102
|
-
end
|
|
2103
|
-
alias :_end_modified :_end_updated
|
|
2104
|
-
alias :_end_dcterms_modified :_end_updated
|
|
2105
|
-
alias :_end_pubdate :_end_updated
|
|
2106
|
-
alias :_end_dc_date :_end_updated
|
|
2107
|
-
|
|
2108
|
-
def _start_created(attrsD)
|
|
2109
|
-
push('created', true)
|
|
2110
|
-
end
|
|
2111
|
-
alias :_start_dcterms_created :_start_created
|
|
2112
|
-
|
|
2113
|
-
def _end_created
|
|
2114
|
-
value = pop('created')
|
|
2115
|
-
_save('created_parsed', parse_date(value))
|
|
2116
|
-
end
|
|
2117
|
-
alias :_end_dcterms_created :_end_created
|
|
2118
|
-
|
|
2119
|
-
def _start_expirationdate(attrsD)
|
|
2120
|
-
push('expired', true)
|
|
2121
|
-
end
|
|
2122
|
-
def _end_expirationdate
|
|
2123
|
-
_save('expired_parsed', parse_date(pop('expired')))
|
|
2124
|
-
end
|
|
2125
|
-
|
|
2126
|
-
def _start_cc_license(attrsD)
|
|
2127
|
-
push('license', true)
|
|
2128
|
-
value = getAttribute(attrsD, 'rdf:resource')
|
|
2129
|
-
if value and not value.empty?
|
|
2130
|
-
elementstack[-1][2] << value
|
|
2131
|
-
pop('license')
|
|
2132
|
-
end
|
|
2133
|
-
end
|
|
2134
|
-
|
|
2135
|
-
def _start_creativecommons_license(attrsD)
|
|
2136
|
-
push('license', true)
|
|
2137
|
-
end
|
|
2138
|
-
|
|
2139
|
-
def _end_creativecommons_license
|
|
2140
|
-
pop('license')
|
|
2141
|
-
end
|
|
2142
|
-
|
|
2143
|
-
def addTag(term, scheme, label)
|
|
2144
|
-
context = getContext()
|
|
2145
|
-
context['tags'] ||= []
|
|
2146
|
-
tags = context['tags']
|
|
2147
|
-
if (term.nil? or term.empty?) and (scheme.nil? or scheme.empty?) and (label.nil? or label.empty?)
|
|
2148
|
-
return
|
|
2149
|
-
end
|
|
2150
|
-
value = FeedParserDict.new({'term' => term, 'scheme' => scheme, 'label' => label})
|
|
2151
|
-
if not tags.include?value
|
|
2152
|
-
context['tags'] << FeedParserDict.new({'term' => term, 'scheme' => scheme, 'label' => label})
|
|
2153
|
-
end
|
|
2154
|
-
end
|
|
2155
|
-
|
|
2156
|
-
def _start_category(attrsD)
|
|
2157
|
-
$stderr << "entering _start_category with #{attrsD}\n" if $debug
|
|
2158
|
-
|
|
2159
|
-
term = attrsD['term']
|
|
2160
|
-
scheme = attrsD['scheme'] || attrsD['domain']
|
|
2161
|
-
label = attrsD['label']
|
|
2162
|
-
addTag(term, scheme, label)
|
|
2163
|
-
push('category', true)
|
|
2164
|
-
end
|
|
2165
|
-
alias :_start_dc_subject :_start_category
|
|
2166
|
-
alias :_start_keywords :_start_category
|
|
2167
|
-
|
|
2168
|
-
def _end_itunes_keywords
|
|
2169
|
-
pop('itunes_keywords').split.each do |term|
|
|
2170
|
-
addTag(term, 'http://www.itunes.com/', nil)
|
|
2171
|
-
end
|
|
2172
|
-
end
|
|
2173
|
-
|
|
2174
|
-
def _start_itunes_category(attrsD)
|
|
2175
|
-
addTag(attrsD['text'], 'http://www.itunes.com/', nil)
|
|
2176
|
-
push('category', true)
|
|
2177
|
-
end
|
|
2178
|
-
|
|
2179
|
-
def _end_category
|
|
2180
|
-
value = pop('category')
|
|
2181
|
-
return if value.nil? or value.empty?
|
|
2182
|
-
context = getContext()
|
|
2183
|
-
tags = context['tags']
|
|
2184
|
-
if value and not value.empty? and not tags.empty? and not tags[-1]['term']:
|
|
2185
|
-
tags[-1]['term'] = value
|
|
2186
|
-
else
|
|
2187
|
-
addTag(value, nil, nil)
|
|
2188
|
-
end
|
|
2189
|
-
end
|
|
2190
|
-
alias :_end_dc_subject :_end_category
|
|
2191
|
-
alias :_end_keywords :_end_category
|
|
2192
|
-
alias :_end_itunes_category :_end_category
|
|
2193
|
-
|
|
2194
|
-
def _start_cloud(attrsD)
|
|
2195
|
-
getContext()['cloud'] = FeedParserDict.new(attrsD)
|
|
2196
|
-
end
|
|
2197
|
-
|
|
2198
|
-
def _start_link(attrsD)
|
|
2199
|
-
attrsD['rel'] ||= 'alternate'
|
|
2200
|
-
attrsD['type'] ||= 'text/html'
|
|
2201
|
-
attrsD = itsAnHrefDamnIt(attrsD)
|
|
2202
|
-
if attrsD.has_key? 'href'
|
|
2203
|
-
attrsD['href'] = resolveURI(attrsD['href'])
|
|
2204
|
-
end
|
|
2205
|
-
expectingText = @infeed || @inentry || @insource
|
|
2206
|
-
context = getContext()
|
|
2207
|
-
context['links'] ||= []
|
|
2208
|
-
context['links'] << FeedParserDict.new(attrsD)
|
|
2209
|
-
if attrsD['rel'] == 'enclosure'
|
|
2210
|
-
_start_enclosure(attrsD)
|
|
2211
|
-
end
|
|
2212
|
-
if attrsD.has_key? 'href'
|
|
2213
|
-
expectingText = false
|
|
2214
|
-
if (attrsD['rel'] == 'alternate') and @html_types.include?mapContentType(attrsD['type'])
|
|
2215
|
-
context['link'] = attrsD['href']
|
|
2216
|
-
end
|
|
2217
|
-
else
|
|
2218
|
-
push('link', expectingText)
|
|
2219
|
-
end
|
|
2220
|
-
end
|
|
2221
|
-
alias :_start_producturl :_start_link
|
|
2222
|
-
|
|
2223
|
-
def _end_link
|
|
2224
|
-
value = pop('link')
|
|
2225
|
-
context = getContext()
|
|
2226
|
-
if @intextinput
|
|
2227
|
-
context['textinput']['link'] = value
|
|
2228
|
-
end
|
|
2229
|
-
if @inimage
|
|
2230
|
-
context['image']['link'] = value
|
|
2231
|
-
end
|
|
2232
|
-
end
|
|
2233
|
-
alias :_end_producturl :_end_link
|
|
2234
|
-
|
|
2235
|
-
def _start_guid(attrsD)
|
|
2236
|
-
@guidislink = ((attrsD['ispermalink'] || 'true') == 'true')
|
|
2237
|
-
push('id', true)
|
|
2238
|
-
end
|
|
2239
|
-
|
|
2240
|
-
def _end_guid
|
|
2241
|
-
value = pop('id')
|
|
2242
|
-
_save('guidislink', (@guidislink and not getContext().has_key?('link')))
|
|
2243
|
-
if @guidislink:
|
|
2244
|
-
# guid acts as link, but only if 'ispermalink' is not present or is 'true',
|
|
2245
|
-
# and only if the item doesn't already have a link element
|
|
2246
|
-
_save('link', value)
|
|
2247
|
-
end
|
|
2248
|
-
end
|
|
2249
|
-
|
|
2250
|
-
|
|
2251
|
-
def _start_title(attrsD)
|
|
2252
|
-
pushContent('title', attrsD, 'text/plain', @infeed || @inentry || @insource)
|
|
2253
|
-
end
|
|
2254
|
-
alias :_start_dc_title :_start_title
|
|
2255
|
-
alias :_start_media_title :_start_title
|
|
2256
|
-
|
|
2257
|
-
def _end_title
|
|
2258
|
-
value = popContent('title')
|
|
2259
|
-
context = getContext()
|
|
2260
|
-
if @intextinput
|
|
2261
|
-
context['textinput']['title'] = value
|
|
2262
|
-
elsif @inimage
|
|
2263
|
-
context['image']['title'] = value
|
|
2264
|
-
end
|
|
2265
|
-
end
|
|
2266
|
-
alias :_end_dc_title :_end_title
|
|
2267
|
-
alias :_end_media_title :_end_title
|
|
2268
|
-
|
|
2269
|
-
def _start_description(attrsD)
|
|
2270
|
-
context = getContext()
|
|
2271
|
-
if context.has_key?('summary')
|
|
2272
|
-
@summaryKey = 'content'
|
|
2273
|
-
_start_content(attrsD)
|
|
2274
|
-
else
|
|
2275
|
-
pushContent('description', attrsD, 'text/html', @infeed || @inentry || @insource)
|
|
2276
|
-
end
|
|
2277
|
-
end
|
|
2278
|
-
|
|
2279
|
-
def _start_abstract(attrsD)
|
|
2280
|
-
pushContent('description', attrsD, 'text/plain', @infeed || @inentry || @insource)
|
|
2281
|
-
end
|
|
2282
|
-
|
|
2283
|
-
def _end_description
|
|
2284
|
-
if @summaryKey == 'content'
|
|
2285
|
-
_end_content()
|
|
2286
|
-
else
|
|
2287
|
-
value = popContent('description')
|
|
2288
|
-
context = getContext()
|
|
2289
|
-
if @intextinput
|
|
2290
|
-
context['textinput']['description'] = value
|
|
2291
|
-
elsif @inimage:
|
|
2292
|
-
context['image']['description'] = value
|
|
2293
|
-
end
|
|
2294
|
-
end
|
|
2295
|
-
@summaryKey = nil
|
|
2296
|
-
end
|
|
2297
|
-
alias :_end_abstract :_end_description
|
|
2298
|
-
|
|
2299
|
-
def _start_info(attrsD)
|
|
2300
|
-
pushContent('info', attrsD, 'text/plain', true)
|
|
2301
|
-
end
|
|
2302
|
-
alias :_start_feedburner_browserfriendly :_start_info
|
|
2303
|
-
|
|
2304
|
-
def _end_info
|
|
2305
|
-
popContent('info')
|
|
2306
|
-
end
|
|
2307
|
-
alias :_end_feedburner_browserfriendly :_end_info
|
|
2308
|
-
|
|
2309
|
-
def _start_generator(attrsD)
|
|
2310
|
-
if attrsD and not attrsD.empty?
|
|
2311
|
-
attrsD = itsAnHrefDamnIt(attrsD)
|
|
2312
|
-
if attrsD.has_key?('href')
|
|
2313
|
-
attrsD['href'] = resolveURI(attrsD['href'])
|
|
2314
|
-
end
|
|
2315
|
-
end
|
|
2316
|
-
getContext()['generator_detail'] = FeedParserDict.new(attrsD)
|
|
2317
|
-
push('generator', true)
|
|
2318
|
-
end
|
|
2319
|
-
|
|
2320
|
-
def _end_generator
|
|
2321
|
-
value = pop('generator')
|
|
2322
|
-
context = getContext()
|
|
2323
|
-
if context.has_key?('generator_detail')
|
|
2324
|
-
context['generator_detail']['name'] = value
|
|
2325
|
-
end
|
|
2326
|
-
end
|
|
2327
|
-
|
|
2328
|
-
def _start_admin_generatoragent(attrsD)
|
|
2329
|
-
push('generator', true)
|
|
2330
|
-
value = getAttribute(attrsD, 'rdf:resource')
|
|
2331
|
-
if value and not value.empty?
|
|
2332
|
-
elementstack[-1][2] << value
|
|
2333
|
-
end
|
|
2334
|
-
pop('generator')
|
|
2335
|
-
getContext()['generator_detail'] = FeedParserDict.new({'href' => value})
|
|
2336
|
-
end
|
|
2337
|
-
|
|
2338
|
-
def _start_admin_errorreportsto(attrsD)
|
|
2339
|
-
push('errorreportsto', true)
|
|
2340
|
-
value = getAttribute(attrsD, 'rdf:resource')
|
|
2341
|
-
if value and not value.empty?
|
|
2342
|
-
@elementstack[-1][2] << value
|
|
2343
|
-
end
|
|
2344
|
-
pop('errorreportsto')
|
|
2345
|
-
end
|
|
2346
|
-
|
|
2347
|
-
def _start_summary(attrsD)
|
|
2348
|
-
context = getContext()
|
|
2349
|
-
if context.has_key?'summary'
|
|
2350
|
-
@summaryKey = 'content'
|
|
2351
|
-
_start_content(attrsD)
|
|
2352
|
-
else
|
|
2353
|
-
@summaryKey = 'summary'
|
|
2354
|
-
pushContent(@summaryKey, attrsD, 'text/plain', true)
|
|
2355
|
-
end
|
|
2356
|
-
end
|
|
2357
|
-
alias :_start_itunes_summary :_start_summary
|
|
2358
|
-
|
|
2359
|
-
def _end_summary
|
|
2360
|
-
if @summaryKey == 'content':
|
|
2361
|
-
_end_content()
|
|
2362
|
-
else
|
|
2363
|
-
popContent(@summaryKey || 'summary')
|
|
2364
|
-
end
|
|
2365
|
-
@summaryKey = nil
|
|
2366
|
-
end
|
|
2367
|
-
alias :_end_itunes_summary :_end_summary
|
|
2368
|
-
|
|
2369
|
-
def _start_enclosure(attrsD)
|
|
2370
|
-
attrsD = itsAnHrefDamnIt(attrsD)
|
|
2371
|
-
getContext()['enclosures'] ||= []
|
|
2372
|
-
getContext()['enclosures'] << FeedParserDict.new(attrsD)
|
|
2373
|
-
href = attrsD['href']
|
|
2374
|
-
if href and not href.empty?
|
|
2375
|
-
context = getContext()
|
|
2376
|
-
if not context['id']
|
|
2377
|
-
context['id'] = href
|
|
2378
|
-
end
|
|
2379
|
-
end
|
|
2380
|
-
end
|
|
2381
|
-
|
|
2382
|
-
def _start_source(attrsD)
|
|
2383
|
-
@insource = true
|
|
2384
|
-
end
|
|
2385
|
-
|
|
2386
|
-
def _end_source
|
|
2387
|
-
@insource = false
|
|
2388
|
-
getContext()['source'] = Marshal.load(Marshal.dump(@sourcedata))
|
|
2389
|
-
@sourcedata.clear()
|
|
2390
|
-
end
|
|
2391
|
-
|
|
2392
|
-
def _start_content(attrsD)
|
|
2393
|
-
pushContent('content', attrsD, 'text/plain', true)
|
|
2394
|
-
src = attrsD['src']
|
|
2395
|
-
if src and not src.empty?:
|
|
2396
|
-
@contentparams['src'] = src
|
|
2397
|
-
end
|
|
2398
|
-
push('content', true)
|
|
2399
|
-
end
|
|
2400
|
-
|
|
2401
|
-
def _start_prodlink(attrsD)
|
|
2402
|
-
pushContent('content', attrsD, 'text/html', true)
|
|
2403
|
-
end
|
|
2404
|
-
|
|
2405
|
-
def _start_body(attrsD)
|
|
2406
|
-
pushContent('content', attrsD, 'application/xhtml+xml', true)
|
|
2407
|
-
end
|
|
2408
|
-
alias :_start_xhtml_body :_start_body
|
|
2409
|
-
|
|
2410
|
-
def _start_content_encoded(attrsD)
|
|
2411
|
-
pushContent('content', attrsD, 'text/html', true)
|
|
2412
|
-
end
|
|
2413
|
-
alias :_start_fullitem :_start_content_encoded
|
|
2414
|
-
|
|
2415
|
-
def _end_content
|
|
2416
|
-
copyToDescription = (['text/plain'] + @html_types).include? mapContentType(@contentparams['type'])
|
|
2417
|
-
value = popContent('content')
|
|
2418
|
-
if copyToDescription
|
|
2419
|
-
_save('description', value)
|
|
2420
|
-
end
|
|
2421
|
-
alias :_end_body :_end_content
|
|
2422
|
-
alias :_end_xhtml_body :_end_content
|
|
2423
|
-
alias :_end_content_encoded :_end_content
|
|
2424
|
-
alias :_end_fullitem :_end_content
|
|
2425
|
-
alias :_end_prodlink :_end_content
|
|
2426
|
-
end
|
|
2427
|
-
|
|
2428
|
-
def _start_itunes_image(attrsD)
|
|
2429
|
-
push('itunes_image', false)
|
|
2430
|
-
getContext()['image'] = FeedParserDict.new({'href' => attrsD['href']})
|
|
2431
|
-
end
|
|
2432
|
-
alias :_start_itunes_link :_start_itunes_image
|
|
2433
|
-
|
|
2434
|
-
def _end_itunes_block
|
|
2435
|
-
value = pop('itunes_block', false)
|
|
2436
|
-
getContext()['itunes_block'] = (value == 'yes') and true or false
|
|
2437
|
-
end
|
|
2438
|
-
|
|
2439
|
-
def _end_itunes_explicit
|
|
2440
|
-
value = pop('itunes_explicit', false)
|
|
2441
|
-
getContext()['itunes_explicit'] = (value == 'yes') and true or false
|
|
2442
|
-
end
|
|
2443
|
-
|
|
2444
|
-
|
|
2445
|
-
# ISO-8601 date parsing routines written by Fazal Majid.
|
|
2446
|
-
# The ISO 8601 standard is very convoluted and irregular - a full ISO 8601
|
|
2447
|
-
# parser is beyond the scope of feedparser and the current Time.iso8601
|
|
2448
|
-
# method does not work.
|
|
2449
|
-
# A single regular expression cannot parse ISO 8601 date formats into groups
|
|
2450
|
-
# as the standard is highly irregular (for instance is 030104 2003-01-04 or
|
|
2451
|
-
# 0301-04-01), so we use templates instead.
|
|
2452
|
-
# Please note the order in templates is significant because we need a
|
|
2453
|
-
# greedy match.
|
|
2454
|
-
def _parse_date_iso8601(dateString)
|
|
2455
|
-
# Parse a variety of ISO-8601-compatible formats like 20040105
|
|
2456
|
-
|
|
2457
|
-
# What I'm about to show you may be the ugliest code in all of
|
|
2458
|
-
# rfeedparser.
|
|
2459
|
-
# FIXME The century regexp maybe not work ('\d\d$' says "two numbers at
|
|
2460
|
-
# end of line" but we then attach more of a regexp.
|
|
2461
|
-
iso8601_regexps = [ '^(\d{4})-?([01]\d)-([0123]\d)',
|
|
2462
|
-
'^(\d{4})-([01]\d)',
|
|
2463
|
-
'^(\d{4})-?([0123]\d\d)',
|
|
2464
|
-
'^(\d\d)-?([01]\d)-?([0123]\d)',
|
|
2465
|
-
'^(\d\d)-?([0123]\d\d)',
|
|
2466
|
-
'^(\d{4})',
|
|
2467
|
-
'-(\d\d)-?([01]\d)',
|
|
2468
|
-
'-([0123]\d\d)',
|
|
2469
|
-
'-(\d\d)',
|
|
2470
|
-
'--([01]\d)-?([0123]\d)',
|
|
2471
|
-
'--([01]\d)',
|
|
2472
|
-
'---([0123]\d)',
|
|
2473
|
-
'(\d\d$)',
|
|
2474
|
-
''
|
|
2475
|
-
]
|
|
2476
|
-
iso8601_values = { '^(\d{4})-?([01]\d)-([0123]\d)' => ['year', 'month', 'day'],
|
|
2477
|
-
'^(\d{4})-([01]\d)' => ['year','month'],
|
|
2478
|
-
'^(\d{4})-?([0123]\d\d)' => ['year', 'ordinal'],
|
|
2479
|
-
'^(\d\d)-?([01]\d)-?([0123]\d)' => ['year','month','day'],
|
|
2480
|
-
'^(\d\d)-?([0123]\d\d)' => ['year','ordinal'],
|
|
2481
|
-
'^(\d{4})' => ['year'],
|
|
2482
|
-
'-(\d\d)-?([01]\d)' => ['year','month'],
|
|
2483
|
-
'-([0123]\d\d)' => ['ordinal'],
|
|
2484
|
-
'-(\d\d)' => ['year'],
|
|
2485
|
-
'--([01]\d)-?([0123]\d)' => ['month','day'],
|
|
2486
|
-
'--([01]\d)' => ['month'],
|
|
2487
|
-
'---([0123]\d)' => ['day'],
|
|
2488
|
-
'(\d\d$)' => ['century'],
|
|
2489
|
-
'' => []
|
|
2490
|
-
}
|
|
2491
|
-
add_to_all = '(T?(\d\d):(\d\d)(?::(\d\d))?([+-](\d\d)(?::(\d\d))?|Z)?)?'
|
|
2492
|
-
add_to_all_fields = ['hour', 'minute', 'second', 'tz', 'tzhour', 'tzmin']
|
|
2493
|
-
# NOTE We use '(?:' to prevent grouping of optional matches (ones trailed
|
|
2494
|
-
# by '?'). The second ':' *are* matched.
|
|
2495
|
-
m = nil
|
|
2496
|
-
param_keys = []
|
|
2497
|
-
iso8601_regexps.each do |s|
|
|
2498
|
-
$stderr << "Trying iso8601 regexp: #{s+add_to_all}\n" if $debug
|
|
2499
|
-
param_keys = iso8601_values[s] + add_to_all_fields
|
|
2500
|
-
m = dateString.match(Regexp.new(s+add_to_all))
|
|
2501
|
-
break if m
|
|
2502
|
-
end
|
|
2503
|
-
return if m.nil? or (m.begin(0).zero? and m.end(0).zero?)
|
|
2504
|
-
|
|
2505
|
-
param_values = m.to_a
|
|
2506
|
-
param_values = param_values[1..-1]
|
|
2507
|
-
params = {}
|
|
2508
|
-
param_keys.each_with_index do |key,i|
|
|
2509
|
-
params[key] = param_values[i]
|
|
2510
|
-
end
|
|
2511
|
-
|
|
2512
|
-
ordinal = params['ordinal'].to_i unless params['ordinal'].nil?
|
|
2513
|
-
year = params['year'] || '--'
|
|
2514
|
-
if year.nil? or year.empty? or year == '--' # FIXME When could the regexp ever return a year equal to '--'?
|
|
2515
|
-
year = Time.now.utc.year
|
|
2516
|
-
elsif year.length == 2
|
|
2517
|
-
# ISO 8601 assumes current century, i.e. 93 -> 2093, NOT 1993
|
|
2518
|
-
year = 100 * (Time.now.utc.year / 100) + year.to_i
|
|
2519
|
-
else
|
|
2520
|
-
year = year.to_i
|
|
2521
|
-
end
|
|
2522
|
-
|
|
2523
|
-
month = params['month'] || '-'
|
|
2524
|
-
if month.nil? or month.empty? or month == '-'
|
|
2525
|
-
# ordinals are NOT normalized by mktime, we simulate them
|
|
2526
|
-
# by setting month=1, day=ordinal
|
|
2527
|
-
if ordinal
|
|
2528
|
-
month = DateTime.ordinal(year,ordinal).month
|
|
2529
|
-
else
|
|
2530
|
-
month = Time.now.utc.month
|
|
2531
|
-
end
|
|
2532
|
-
end
|
|
2533
|
-
month = month.to_i unless month.nil?
|
|
2534
|
-
day = params['day']
|
|
2535
|
-
if day.nil? or day.empty?
|
|
2536
|
-
# see above
|
|
2537
|
-
if ordinal
|
|
2538
|
-
day = DateTime.ordinal(year,ordinal).day
|
|
2539
|
-
elsif params['century'] or params['year'] or params['month']
|
|
2540
|
-
day = 1
|
|
2541
|
-
else
|
|
2542
|
-
day = Time.now.utc.day
|
|
2543
|
-
end
|
|
2544
|
-
else
|
|
2545
|
-
day = day.to_i
|
|
2546
|
-
end
|
|
2547
|
-
# special case of the century - is the first year of the 21st century
|
|
2548
|
-
# 2000 or 2001 ? The debate goes on...
|
|
2549
|
-
if params.has_key? 'century'
|
|
2550
|
-
year = (params['century'].to_i - 1) * 100 + 1
|
|
2551
|
-
end
|
|
2552
|
-
# in ISO 8601 most fields are optional
|
|
2553
|
-
hour = params['hour'].to_i
|
|
2554
|
-
minute = params['minute'].to_i
|
|
2555
|
-
second = params['second'].to_i
|
|
2556
|
-
weekday = nil
|
|
2557
|
-
# daylight savings is complex, but not needed for feedparser's purposes
|
|
2558
|
-
# as time zones, if specified, include mention of whether it is active
|
|
2559
|
-
# (e.g. PST vs. PDT, CET). Using -1 is implementation-dependent and
|
|
2560
|
-
# and most implementations have DST bugs
|
|
2561
|
-
tm = [second, minute, hour, day, month, year, nil, ordinal, false, nil]
|
|
2562
|
-
tz = params['tz']
|
|
2563
|
-
if tz and not tz.empty? and tz != 'Z'
|
|
2564
|
-
# FIXME does this cross over days?
|
|
2565
|
-
if tz[0] == '-'
|
|
2566
|
-
tm[3] += params['tzhour'].to_i
|
|
2567
|
-
tm[4] += params['tzmin'].to_i
|
|
2568
|
-
elsif tz[0] == '+'
|
|
2569
|
-
tm[3] -= params['tzhour'].to_i
|
|
2570
|
-
tm[4] -= params['tzmin'].to_i
|
|
2571
|
-
else
|
|
2572
|
-
return nil
|
|
2573
|
-
end
|
|
2574
|
-
end
|
|
2575
|
-
return Time.utc(*tm) # Magic!
|
|
2576
|
-
|
|
2577
|
-
end
|
|
2578
|
-
|
|
2579
|
-
def _parse_date_onblog(dateString)
|
|
2580
|
-
# Parse a string according to the OnBlog 8-bit date format
|
|
2581
|
-
# 8-bit date handling routes written by ytrewq1
|
|
2582
|
-
korean_year = u("년") # b3e2 in euc-kr
|
|
2583
|
-
korean_month = u("월") # bff9 in euc-kr
|
|
2584
|
-
korean_day = u("일") # c0cf in euc-kr
|
|
2585
|
-
|
|
2586
|
-
|
|
2587
|
-
korean_onblog_date_re = /(\d{4})#{korean_year}\s+(\d{2})#{korean_month}\s+(\d{2})#{korean_day}\s+(\d{2}):(\d{2}):(\d{2})/
|
|
2588
|
-
|
|
2589
|
-
|
|
2590
|
-
m = korean_onblog_date_re.match(dateString)
|
|
2591
|
-
return unless m
|
|
2592
|
-
w3dtfdate = "#{m[1]}-#{m[2]}-#{m[3]}T#{m[4]}:#{m[5]}:#{m[6]}+09:00"
|
|
2593
|
-
|
|
2594
|
-
$stderr << "OnBlog date parsed as: %s\n" % w3dtfdate if $debug
|
|
2595
|
-
return _parse_date_w3dtf(w3dtfdate)
|
|
2596
|
-
end
|
|
2597
|
-
|
|
2598
|
-
def _parse_date_nate(dateString)
|
|
2599
|
-
# Parse a string according to the Nate 8-bit date format
|
|
2600
|
-
# 8-bit date handling routes written by ytrewq1
|
|
2601
|
-
korean_am = u("오전") # bfc0 c0fc in euc-kr
|
|
2602
|
-
korean_pm = u("오후") # bfc0 c8c4 in euc-kr
|
|
2603
|
-
|
|
2604
|
-
korean_nate_date_re = /(\d{4})-(\d{2})-(\d{2})\s+(#{korean_am}|#{korean_pm})\s+(\d{0,2}):(\d{0,2}):(\d{0,2})/
|
|
2605
|
-
m = korean_nate_date_re.match(dateString)
|
|
2606
|
-
return unless m
|
|
2607
|
-
hour = m[5].to_i
|
|
2608
|
-
ampm = m[4]
|
|
2609
|
-
if ampm == korean_pm
|
|
2610
|
-
hour += 12
|
|
2611
|
-
end
|
|
2612
|
-
hour = hour.to_s.rjust(2,'0')
|
|
2613
|
-
w3dtfdate = "#{m[1]}-#{m[2]}-#{m[3]}T#{hour}:#{m[6]}:#{m[7]}+09:00"
|
|
2614
|
-
$stderr << "Nate date parsed as: %s\n" % w3dtfdate if $debug
|
|
2615
|
-
return _parse_date_w3dtf(w3dtfdate)
|
|
2616
|
-
end
|
|
2617
|
-
|
|
2618
|
-
def _parse_date_mssql(dateString)
|
|
2619
|
-
mssql_date_re = /(\d{4})-(\d{2})-(\d{2})\s+(\d{2}):(\d{2}):(\d{2})(\.\d+)?/
|
|
2620
|
-
|
|
2621
|
-
m = mssql_date_re.match(dateString)
|
|
2622
|
-
return unless m
|
|
2623
|
-
w3dtfdate = "#{m[1]}-#{m[2]}-#{m[3]}T#{m[4]}:#{m[5]}:#{m[6]}+09:00"
|
|
2624
|
-
$stderr << "MS SQL date parsed as: %s\n" % w3dtfdate if $debug
|
|
2625
|
-
return _parse_date_w3dtf(w3dtfdate)
|
|
2626
|
-
end
|
|
2627
|
-
|
|
2628
|
-
def _parse_date_greek(dateString)
|
|
2629
|
-
# Parse a string according to a Greek 8-bit date format
|
|
2630
|
-
# Unicode strings for Greek date strings
|
|
2631
|
-
greek_months = {
|
|
2632
|
-
u("Ιαν") => u("Jan"), # c9e1ed in iso-8859-7
|
|
2633
|
-
u("Φεβ") => u("Feb"), # d6e5e2 in iso-8859-7
|
|
2634
|
-
u("Μάώ") => u("Mar"), # ccdcfe in iso-8859-7
|
|
2635
|
-
u("Μαώ") => u("Mar"), # cce1fe in iso-8859-7
|
|
2636
|
-
u("Απρ") => u("Apr"), # c1f0f1 in iso-8859-7
|
|
2637
|
-
u("Μάι") => u("May"), # ccdce9 in iso-8859-7
|
|
2638
|
-
u("Μαϊ") => u("May"), # cce1fa in iso-8859-7
|
|
2639
|
-
u("Μαι") => u("May"), # cce1e9 in iso-8859-7
|
|
2640
|
-
u("Ιούν") => u("Jun"), # c9effded in iso-8859-7
|
|
2641
|
-
u("Ιον") => u("Jun"), # c9efed in iso-8859-7
|
|
2642
|
-
u("Ιούλ") => u("Jul"), # c9effdeb in iso-8859-7
|
|
2643
|
-
u("Ιολ") => u("Jul"), # c9f9eb in iso-8859-7
|
|
2644
|
-
u("Αύγ") => u("Aug"), # c1fde3 in iso-8859-7
|
|
2645
|
-
u("Αυγ") => u("Aug"), # c1f5e3 in iso-8859-7
|
|
2646
|
-
u("Σεπ") => u("Sep"), # d3e5f0 in iso-8859-7
|
|
2647
|
-
u("Οκτ") => u("Oct"), # cfeaf4 in iso-8859-7
|
|
2648
|
-
u("Νοέ") => u("Nov"), # cdefdd in iso-8859-7
|
|
2649
|
-
u("Νοε") => u("Nov"), # cdefe5 in iso-8859-7
|
|
2650
|
-
u("Δεκ") => u("Dec"), # c4e5ea in iso-8859-7
|
|
2651
|
-
}
|
|
2652
|
-
|
|
2653
|
-
greek_wdays = {
|
|
2654
|
-
u("Κυρ") => u("Sun"), # caf5f1 in iso-8859-7
|
|
2655
|
-
u("Δευ") => u("Mon"), # c4e5f5 in iso-8859-7
|
|
2656
|
-
u("Τρι") => u("Tue"), # d4f1e9 in iso-8859-7
|
|
2657
|
-
u("Τετ") => u("Wed"), # d4e5f4 in iso-8859-7
|
|
2658
|
-
u("Πεμ") => u("Thu"), # d0e5ec in iso-8859-7
|
|
2659
|
-
u("Παρ") => u("Fri"), # d0e1f1 in iso-8859-7
|
|
2660
|
-
u("Σαβ") => u("Sat"), # d3e1e2 in iso-8859-7
|
|
2661
|
-
}
|
|
2662
|
-
|
|
2663
|
-
greek_date_format = /([^,]+),\s+(\d{2})\s+([^\s]+)\s+(\d{4})\s+(\d{2}):(\d{2}):(\d{2})\s+([^\s]+)/
|
|
2664
|
-
|
|
2665
|
-
m = greek_date_format.match(dateString)
|
|
2666
|
-
return unless m
|
|
2667
|
-
begin
|
|
2668
|
-
wday = greek_wdays[m[1]]
|
|
2669
|
-
month = greek_months[m[3]]
|
|
2670
|
-
rescue
|
|
2671
|
-
return nil
|
|
2672
|
-
end
|
|
2673
|
-
rfc822date = "#{wday}, #{m[2]} #{month} #{m[4]} #{m[5]}:#{m[6]}:#{m[7]} #{m[8]}"
|
|
2674
|
-
$stderr << "Greek date parsed as: #{rfc822date}\n" if $debug
|
|
2675
|
-
return _parse_date_rfc822(rfc822date)
|
|
2676
|
-
end
|
|
2677
|
-
|
|
2678
|
-
def _parse_date_hungarian(dateString)
|
|
2679
|
-
# Parse a string according to a Hungarian 8-bit date format.
|
|
2680
|
-
hungarian_date_format_re = /(\d{4})-([^-]+)-(\d{0,2})T(\d{0,2}):(\d{2})((\+|-)(\d{0,2}:\d{2}))/
|
|
2681
|
-
m = hungarian_date_format_re.match(dateString)
|
|
2682
|
-
return unless m
|
|
2683
|
-
|
|
2684
|
-
# Unicode strings for Hungarian date strings
|
|
2685
|
-
hungarian_months = {
|
|
2686
|
-
u("január") => u("01"), # e1 in iso-8859-2
|
|
2687
|
-
u("februári") => u("02"), # e1 in iso-8859-2
|
|
2688
|
-
u("március") => u("03"), # e1 in iso-8859-2
|
|
2689
|
-
u("április") => u("04"), # e1 in iso-8859-2
|
|
2690
|
-
u("máujus") => u("05"), # e1 in iso-8859-2
|
|
2691
|
-
u("június") => u("06"), # fa in iso-8859-2
|
|
2692
|
-
u("július") => u("07"), # fa in iso-8859-2
|
|
2693
|
-
u("augusztus") => u("08"),
|
|
2694
|
-
u("szeptember") => u("09"),
|
|
2695
|
-
u("október") => u("10"), # f3 in iso-8859-2
|
|
2696
|
-
u("november") => u("11"),
|
|
2697
|
-
u("december") => u("12"),
|
|
2698
|
-
}
|
|
2699
|
-
begin
|
|
2700
|
-
month = hungarian_months[m[2]]
|
|
2701
|
-
day = m[3].rjust(2,'0')
|
|
2702
|
-
hour = m[4].rjust(2,'0')
|
|
2703
|
-
rescue
|
|
2704
|
-
return
|
|
2705
|
-
end
|
|
2706
|
-
|
|
2707
|
-
w3dtfdate = "#{m[1]}-#{month}-#{day}T#{hour}:#{m[5]}:00#{m[6]}"
|
|
2708
|
-
$stderr << "Hungarian date parsed as: #{w3dtfdate}\n" if $debug
|
|
2709
|
-
return _parse_date_w3dtf(w3dtfdate)
|
|
2710
|
-
end
|
|
2711
|
-
|
|
2712
|
-
def rollover(num, modulus)
|
|
2713
|
-
return num % modulus, num / modulus
|
|
2714
|
-
end
|
|
2715
|
-
|
|
2716
|
-
def set_self(num, modulus)
|
|
2717
|
-
r = num / modulus
|
|
2718
|
-
if r == 0
|
|
2719
|
-
return num
|
|
2720
|
-
end
|
|
2721
|
-
return r
|
|
2722
|
-
end
|
|
2723
|
-
# W3DTF-style date parsing
|
|
2724
|
-
# FIXME shouldn't it be "W3CDTF"?
|
|
2725
|
-
def _parse_date_w3dtf(dateString)
|
|
2726
|
-
# Ruby's Time docs claim w3cdtf is an alias for iso8601 which is an alias for xmlschema
|
|
2727
|
-
# Whatever it is, it doesn't work. This has been fixed in Ruby 1.9 and
|
|
2728
|
-
# in Ruby on Rails, but not really. They don't fix the 25 hour or 61 minute or 61 second rollover and fail in other ways.
|
|
2729
|
-
|
|
2730
|
-
m = dateString.match(/^(\d{4})-?(?:(?:([01]\d)-?(?:([0123]\d)(?:T(\d\d):(\d\d):(\d\d)([+-]\d\d:\d\d|Z))?)?)?)?/)
|
|
2731
|
-
|
|
2732
|
-
w3 = m[1..3].map{|s| s=s.to_i; s += 1 if s == 0;s} # Map the year, month and day to integers and, if they were nil, set them to 1
|
|
2733
|
-
w3 += m[4..6].map{|s| s.to_i} # Map the hour, minute and second to integers
|
|
2734
|
-
w3 << m[-1] # Leave the timezone as a String
|
|
2735
|
-
|
|
2736
|
-
# FIXME this next bit needs some serious refactoring
|
|
2737
|
-
# Rollover times. 0 minutes and 61 seconds -> 1 minute and 1 second
|
|
2738
|
-
w3[5],r = rollover(w3[5], 60) # rollover seconds
|
|
2739
|
-
w3[4] += r
|
|
2740
|
-
w3[4],r = rollover(w3[4], 60) # rollover minutes
|
|
2741
|
-
w3[3] += r
|
|
2742
|
-
w3[3],r = rollover(w3[3], 24) # rollover hours
|
|
2743
|
-
|
|
2744
|
-
w3[2] = w3[2] + r
|
|
2745
|
-
if w3[1] > 12
|
|
2746
|
-
w3[1],r = rollover(w3[1],12)
|
|
2747
|
-
w3[1] = 12 if w3[1] == 0
|
|
2748
|
-
w3[0] += r
|
|
2749
|
-
end
|
|
2750
|
-
|
|
2751
|
-
num_days = Time.days_in_month(w3[1], w3[0])
|
|
2752
|
-
while w3[2] > num_days
|
|
2753
|
-
w3[2] -= num_days
|
|
2754
|
-
w3[1] += 1
|
|
2755
|
-
if w3[1] > 12
|
|
2756
|
-
w3[0] += 1
|
|
2757
|
-
w3[1] = set_self(w3[1], 12)
|
|
2758
|
-
end
|
|
2759
|
-
num_days = Time.days_in_month(w3[1], w3[0])
|
|
2760
|
-
end
|
|
2761
|
-
|
|
2762
|
-
|
|
2763
|
-
unless w3[6].class != String
|
|
2764
|
-
if /^-/ =~ w3[6] # Zone offset goes backwards
|
|
2765
|
-
w3[6][0] = '+'
|
|
2766
|
-
elsif /^\+/ =~ w3[6]
|
|
2767
|
-
w3[6][0] = '-'
|
|
2768
|
-
end
|
|
2769
|
-
end
|
|
2770
|
-
return Time.utc(w3[0], w3[1], w3[2] , w3[3], w3[4], w3[5])+Time.zone_offset(w3[6] || "UTC")
|
|
2771
|
-
end
|
|
2772
|
-
|
|
2773
|
-
def _parse_date_rfc822(dateString)
|
|
2774
|
-
# Parse an RFC822, RFC1123, RFC2822 or asctime-style date
|
|
2775
|
-
# These first few lines are to fix up the stupid proprietary format from Disney
|
|
2776
|
-
unknown_timezones = { 'AT' => 'EDT', 'ET' => 'EST',
|
|
2777
|
-
'CT' => 'CST', 'MT' => 'MST',
|
|
2778
|
-
'PT' => 'PST'
|
|
2779
|
-
}
|
|
2780
|
-
|
|
2781
|
-
mon = dateString.split[2]
|
|
2782
|
-
if mon.length > 3 and Time::RFC2822_MONTH_NAME.include?mon[0..2]
|
|
2783
|
-
dateString.sub!(mon,mon[0..2])
|
|
2784
|
-
end
|
|
2785
|
-
if dateString[-3..-1] != "GMT" and unknown_timezones[dateString[-2..-1]]
|
|
2786
|
-
dateString[-2..-1] = unknown_timezones[dateString[-2..-1]]
|
|
2787
|
-
end
|
|
2788
|
-
# Okay, the Disney date format should be fixed up now.
|
|
2789
|
-
rfc = dateString.match(/([A-Za-z]{3}), ([0123]\d) ([A-Za-z]{3}) (\d{4})( (\d\d):(\d\d)(?::(\d\d))? ([A-Za-z]{3}))?/)
|
|
2790
|
-
if rfc.to_a.length > 1 and rfc.to_a.include? nil
|
|
2791
|
-
dow, day, mon, year, hour, min, sec, tz = rfc[1..-1]
|
|
2792
|
-
hour,min,sec = [hour,min,sec].map{|e| e.to_s.rjust(2,'0') }
|
|
2793
|
-
tz ||= "GMT"
|
|
2794
|
-
end
|
|
2795
|
-
asctime_match = dateString.match(/([A-Za-z]{3}) ([A-Za-z]{3}) (\d?\d) (\d\d):(\d\d):(\d\d) ([A-Za-z]{3}) (\d\d\d\d)/).to_a
|
|
2796
|
-
if asctime_match.to_a.length > 1
|
|
2797
|
-
# Month-abbr dayofmonth hour:minute:second year
|
|
2798
|
-
dow, mon, day, hour, min, sec, tz, year = asctime_match[1..-1]
|
|
2799
|
-
day.to_s.rjust(2,'0')
|
|
2800
|
-
end
|
|
2801
|
-
if (rfc.to_a.length > 1 and rfc.to_a.include? nil) or asctime_match.to_a.length > 1
|
|
2802
|
-
ds = "#{dow}, #{day} #{mon} #{year} #{hour}:#{min}:#{sec} #{tz}"
|
|
2803
|
-
else
|
|
2804
|
-
ds = dateString
|
|
2805
|
-
end
|
|
2806
|
-
t = Time.rfc2822(ds).utc
|
|
2807
|
-
return t
|
|
2808
|
-
end
|
|
2809
|
-
|
|
2810
|
-
def _parse_date_perforce(aDateString) # FIXME not in 4.1?
|
|
2811
|
-
# Parse a date in yyyy/mm/dd hh:mm:ss TTT format
|
|
2812
|
-
# Note that there is a day of the week at the beginning
|
|
2813
|
-
# Ex. Fri, 2006/09/15 08:19:53 EDT
|
|
2814
|
-
return Time.parse(aDateString).utc
|
|
2815
|
-
end
|
|
2816
|
-
|
|
2817
|
-
def extract_tuple(atime)
|
|
2818
|
-
# NOTE leave the error handling to parse_date
|
|
2819
|
-
t = [atime.year, atime.month, atime.mday, atime.hour,
|
|
2820
|
-
atime.min, atime.sec, (atime.wday-1) % 7, atime.yday,
|
|
2821
|
-
atime.isdst
|
|
2822
|
-
]
|
|
2823
|
-
# yay for modulus! yaaaaaay! its 530 am and i should be sleeping! yaay!
|
|
2824
|
-
t[0..-2].map!{|s| s.to_i}
|
|
2825
|
-
t[-1] = t[-1] ? 1 : 0
|
|
2826
|
-
return t
|
|
2827
|
-
end
|
|
2828
|
-
|
|
2829
|
-
def parse_date(dateString)
|
|
2830
|
-
@date_handlers.each do |handler|
|
|
2831
|
-
begin
|
|
2832
|
-
$stderr << "Trying date_handler #{handler}\n" if $debug
|
|
2833
|
-
datething = extract_tuple(send(handler,dateString))
|
|
2834
|
-
return datething
|
|
2835
|
-
rescue Exception => e
|
|
2836
|
-
$stderr << "#{handler} raised #{e}\n" if $debug
|
|
2837
|
-
end
|
|
2838
|
-
end
|
|
2839
|
-
return nil
|
|
2840
|
-
end
|
|
2841
|
-
|
|
2842
|
-
end # End FeedParserMixin
|
|
2843
|
-
|
|
2844
|
-
class StrictFeedParser < XML::SAX::HandlerBase # expat
|
|
2845
|
-
include FeedParserMixin
|
|
2846
|
-
|
|
2847
|
-
attr_accessor :bozo, :entries, :feeddata, :exc
|
|
2848
|
-
def initialize(baseuri, baselang, encoding)
|
|
2849
|
-
$stderr << "trying StrictFeedParser\n" if $debug
|
|
2850
|
-
startup(baseuri, baselang, encoding)
|
|
2851
|
-
@bozo = false
|
|
2852
|
-
@exc = nil
|
|
2853
|
-
super()
|
|
2854
|
-
end
|
|
2855
|
-
|
|
2856
|
-
def getPos
|
|
2857
|
-
[@locator.getSystemId, @locator.getLineNumber]
|
|
2858
|
-
end
|
|
2859
|
-
|
|
2860
|
-
def getAttrs(attrs)
|
|
2861
|
-
ret = []
|
|
2862
|
-
for i in 0..attrs.getLength
|
|
2863
|
-
ret.push([attrs.getName(i), attrs.getValue(i)])
|
|
2864
|
-
end
|
|
2865
|
-
ret
|
|
2866
|
-
end
|
|
2867
|
-
|
|
2868
|
-
def setDocumentLocator(loc)
|
|
2869
|
-
@locator = loc
|
|
2870
|
-
end
|
|
2871
|
-
|
|
2872
|
-
def startDoctypeDecl(name, pub_sys, long_name, uri)
|
|
2873
|
-
#Nothing is done here. What could we do that is neat and useful?
|
|
2874
|
-
end
|
|
2875
|
-
|
|
2876
|
-
def startNamespaceDecl(prefix, uri)
|
|
2877
|
-
trackNamespace(prefix, uri)
|
|
2878
|
-
end
|
|
2879
|
-
|
|
2880
|
-
def endNamespaceDecl(prefix)
|
|
2881
|
-
end
|
|
2882
|
-
|
|
2883
|
-
def startElement(name, attrs)
|
|
2884
|
-
name =~ /^(([^;]*);)?(.+)$/ # Snag namespaceuri from name
|
|
2885
|
-
namespaceuri = ($2 || '').downcase
|
|
2886
|
-
name = $3
|
|
2887
|
-
if /backend\.userland\.com\/rss/ =~ namespaceuri
|
|
2888
|
-
# match any backend.userland.com namespace
|
|
2889
|
-
namespaceuri = 'http://backend.userland.com/rss'
|
|
2890
|
-
end
|
|
2891
|
-
prefix = @matchnamespaces[namespaceuri]
|
|
2892
|
-
# No need to raise UndeclaredNamespace, Expat does that for us with
|
|
2893
|
-
"unbound prefix (XMLParserError)"
|
|
2894
|
-
if prefix and not prefix.empty?
|
|
2895
|
-
name = prefix + ':' + name
|
|
2896
|
-
end
|
|
2897
|
-
name.downcase!
|
|
2898
|
-
unknown_starttag(name, attrs)
|
|
2899
|
-
end
|
|
2900
|
-
|
|
2901
|
-
def character(text, start, length)
|
|
2902
|
-
#handle_data(CGI.unescapeHTML(text))
|
|
2903
|
-
handle_data(text)
|
|
2904
|
-
end
|
|
2905
|
-
# expat provides "character" not "characters"!
|
|
2906
|
-
alias :characters :character # Just in case.
|
|
2907
|
-
|
|
2908
|
-
def startCdata(content)
|
|
2909
|
-
handle_data(content)
|
|
2910
|
-
end
|
|
2911
|
-
|
|
2912
|
-
def endElement(name)
|
|
2913
|
-
name =~ /^(([^;]*);)?(.+)$/ # Snag namespaceuri from name
|
|
2914
|
-
namespaceuri = ($2 || '').downcase
|
|
2915
|
-
prefix = @matchnamespaces[namespaceuri]
|
|
2916
|
-
if prefix and not prefix.empty?
|
|
2917
|
-
localname = prefix + ':' + name
|
|
2918
|
-
end
|
|
2919
|
-
name.downcase!
|
|
2920
|
-
unknown_endtag(name)
|
|
2921
|
-
end
|
|
2922
|
-
|
|
2923
|
-
def comment(comment)
|
|
2924
|
-
handle_comment(comment)
|
|
2925
|
-
end
|
|
2926
|
-
|
|
2927
|
-
def entityDecl(*foo)
|
|
2928
|
-
end
|
|
2929
|
-
|
|
2930
|
-
def unparsedEntityDecl(*foo)
|
|
2931
|
-
end
|
|
2932
|
-
def error(exc)
|
|
2933
|
-
@bozo = true
|
|
2934
|
-
@exc = exc
|
|
2935
|
-
end
|
|
2936
|
-
|
|
2937
|
-
def fatalError(exc)
|
|
2938
|
-
error(exc)
|
|
2939
|
-
raise exc
|
|
2940
|
-
end
|
|
2941
|
-
end
|
|
2942
|
-
|
|
2943
|
-
class LooseFeedParser < BetterSGMLParser
|
|
2944
|
-
include FeedParserMixin
|
|
2945
|
-
# We write the methods that were in BaseHTMLProcessor in the python code
|
|
2946
|
-
# in here directly. We do this because if we inherited from
|
|
2947
|
-
# BaseHTMLProcessor but then included from FeedParserMixin, the methods
|
|
2948
|
-
# of Mixin would overwrite the methods we inherited from
|
|
2949
|
-
# BaseHTMLProcessor. This is exactly the opposite of what we want to
|
|
2950
|
-
# happen!
|
|
2951
|
-
|
|
2952
|
-
attr_accessor :encoding, :bozo, :feeddata, :entries, :namespacesInUse
|
|
2953
|
-
|
|
2954
|
-
Elements_No_End_Tag = ['area', 'base', 'basefont', 'br', 'col', 'frame', 'hr',
|
|
2955
|
-
'img', 'input', 'isindex', 'link', 'meta', 'param']
|
|
2956
|
-
New_Declname_Re = /[a-zA-Z][-_.a-zA-Z0-9:]*\s*/
|
|
2957
|
-
alias :sgml_feed :feed # feed needs to mapped to feeddata, not the SGMLParser method feed. I think.
|
|
2958
|
-
def feed
|
|
2959
|
-
@feeddata
|
|
2960
|
-
end
|
|
2961
|
-
def feed=(data)
|
|
2962
|
-
@feeddata = data
|
|
2963
|
-
end
|
|
2964
|
-
|
|
2965
|
-
def initialize(baseuri, baselang, encoding)
|
|
2966
|
-
startup(baseuri, baselang, encoding)
|
|
2967
|
-
super() # Keep the parentheses! No touchy.
|
|
2968
|
-
end
|
|
2969
|
-
|
|
2970
|
-
def reset
|
|
2971
|
-
@pieces = []
|
|
2972
|
-
super
|
|
2973
|
-
end
|
|
2974
|
-
|
|
2975
|
-
def parse(data)
|
|
2976
|
-
data.gsub!(/<!((?!DOCTYPE|--|\[))/i, '<!\1')
|
|
2977
|
-
data.gsub!(/<([^<\s]+?)\s*\/>/) do |tag|
|
|
2978
|
-
clean = tag[1..-3].strip
|
|
2979
|
-
if Elements_No_End_Tag.include?clean
|
|
2980
|
-
tag
|
|
2981
|
-
else
|
|
2982
|
-
'<'+clean+'></'+clean+'>'
|
|
2983
|
-
end
|
|
2984
|
-
end
|
|
2985
|
-
|
|
2986
|
-
data.gsub!(/'/, "'")
|
|
2987
|
-
data.gsub!(/"/, "'")
|
|
2988
|
-
if @encoding and not @encoding.empty? # FIXME unicode check type(u'')
|
|
2989
|
-
data = uconvert(data,'utf-8',@encoding)
|
|
2990
|
-
end
|
|
2991
|
-
sgml_feed(data) # see the alias above
|
|
2992
|
-
end
|
|
2993
|
-
|
|
2994
|
-
|
|
2995
|
-
def decodeEntities(element, data)
|
|
2996
|
-
data.gsub!('<', '<')
|
|
2997
|
-
data.gsub!('<', '<')
|
|
2998
|
-
data.gsub!('>', '>')
|
|
2999
|
-
data.gsub!('>', '>')
|
|
3000
|
-
data.gsub!('&', '&')
|
|
3001
|
-
data.gsub!('&', '&')
|
|
3002
|
-
data.gsub!('"', '"')
|
|
3003
|
-
data.gsub!('"', '"')
|
|
3004
|
-
data.gsub!(''', ''')
|
|
3005
|
-
data.gsub!(''', ''')
|
|
3006
|
-
if @contentparams.has_key? 'type' and not ((@contentparams['type'] || 'xml') =~ /xml$/u)
|
|
3007
|
-
data.gsub!('<', '<')
|
|
3008
|
-
data.gsub!('>', '>')
|
|
3009
|
-
data.gsub!('&', '&')
|
|
3010
|
-
data.gsub!('"', '"')
|
|
3011
|
-
data.gsub!(''', "'")
|
|
3012
|
-
end
|
|
3013
|
-
return data
|
|
3014
|
-
end
|
|
3015
|
-
end
|
|
3016
|
-
|
|
3017
|
-
def FeedParser.resolveRelativeURIs(htmlSource, baseURI, encoding)
|
|
3018
|
-
$stderr << "entering resolveRelativeURIs\n" if $debug # FIXME write a decent logger
|
|
3019
|
-
relative_uris = [ ['a','href'],
|
|
3020
|
-
['applet','codebase'],
|
|
3021
|
-
['area','href'],
|
|
3022
|
-
['blockquote','cite'],
|
|
3023
|
-
['body','background'],
|
|
3024
|
-
['del','cite'],
|
|
3025
|
-
['form','action'],
|
|
3026
|
-
['frame','longdesc'],
|
|
3027
|
-
['frame','src'],
|
|
3028
|
-
['iframe','longdesc'],
|
|
3029
|
-
['iframe','src'],
|
|
3030
|
-
['head','profile'],
|
|
3031
|
-
['img','longdesc'],
|
|
3032
|
-
['img','src'],
|
|
3033
|
-
['img','usemap'],
|
|
3034
|
-
['input','src'],
|
|
3035
|
-
['input','usemap'],
|
|
3036
|
-
['ins','cite'],
|
|
3037
|
-
['link','href'],
|
|
3038
|
-
['object','classid'],
|
|
3039
|
-
['object','codebase'],
|
|
3040
|
-
['object','data'],
|
|
3041
|
-
['object','usemap'],
|
|
3042
|
-
['q','cite'],
|
|
3043
|
-
['script','src'],
|
|
3044
|
-
]
|
|
3045
|
-
h = Hpricot(htmlSource)
|
|
3046
|
-
relative_uris.each do |l|
|
|
3047
|
-
ename, eattr = l
|
|
3048
|
-
h.search(ename).each do |elem|
|
|
3049
|
-
euri = elem.attributes[eattr]
|
|
3050
|
-
if euri and not euri.empty? and URI.parse(URI.encode(euri)).relative?
|
|
3051
|
-
elem.attributes[eattr] = urljoin(baseURI, euri)
|
|
3052
|
-
end
|
|
3053
|
-
end
|
|
3054
|
-
end
|
|
3055
|
-
return h.to_html
|
|
3056
|
-
end
|
|
3057
|
-
|
|
3058
|
-
class SanitizerDoc < Hpricot::Doc
|
|
3059
|
-
|
|
3060
|
-
def scrub
|
|
3061
|
-
traverse_all_element do |e|
|
|
3062
|
-
if e.elem?
|
|
3063
|
-
if Acceptable_Elements.include?e.name
|
|
3064
|
-
e.strip_attributes
|
|
3065
|
-
else
|
|
3066
|
-
if Unacceptable_Elements_With_End_Tag.include?e.name
|
|
3067
|
-
e.inner_html = ''
|
|
3068
|
-
end
|
|
3069
|
-
e.swap(SanitizerDoc.new(e.children).scrub.to_html)
|
|
3070
|
-
# This works because the children swapped in are brought in "after" the current element.
|
|
3071
|
-
end
|
|
3072
|
-
elsif e.doctype?
|
|
3073
|
-
e.parent.children.delete(e)
|
|
3074
|
-
elsif e.text?
|
|
3075
|
-
ets = e.to_s
|
|
3076
|
-
ets.gsub!(/'/, "'")
|
|
3077
|
-
ets.gsub!(/"/, '"')
|
|
3078
|
-
ets.gsub!(/\r/,'')
|
|
3079
|
-
e.swap(ets)
|
|
3080
|
-
else
|
|
3081
|
-
end
|
|
3082
|
-
end
|
|
3083
|
-
# yes, that '/' should be there. It's a search method. See the Hpricot docs.
|
|
3084
|
-
|
|
3085
|
-
unless $compatible # FIXME nonworking
|
|
3086
|
-
(self/tag).strip_style(@config[:allow_css_properties], @config[:allow_css_keywords])
|
|
3087
|
-
end
|
|
3088
|
-
return self
|
|
3089
|
-
end
|
|
3090
|
-
end
|
|
3091
|
-
|
|
3092
|
-
def SanitizerDoc(html)
|
|
3093
|
-
FeedParser::SanitizerDoc.new(Hpricot.make(html))
|
|
3094
|
-
end
|
|
3095
|
-
module_function(:SanitizerDoc)
|
|
3096
|
-
|
|
3097
|
-
def self.sanitizeHTML(html,encoding)
|
|
3098
|
-
# FIXME Tidy not yet supported
|
|
3099
|
-
html = html.gsub(/<!((?!DOCTYPE|--|\[))/, '<!\1')
|
|
3100
|
-
h = SanitizerDoc(html)
|
|
3101
|
-
h = h.scrub
|
|
3102
|
-
return h.to_html.strip
|
|
3103
|
-
end
|
|
3104
|
-
|
|
3105
|
-
|
|
3106
|
-
|
|
3107
|
-
def self.getCharacterEncoding(feed, xml_data)
|
|
3108
|
-
# Get the character encoding of the XML document
|
|
3109
|
-
$stderr << "In getCharacterEncoding\n" if $debug
|
|
3110
|
-
sniffed_xml_encoding = nil
|
|
3111
|
-
xml_encoding = nil
|
|
3112
|
-
true_encoding = nil
|
|
3113
|
-
begin
|
|
3114
|
-
http_headers = feed.meta
|
|
3115
|
-
http_content_type = feed.meta['content-type'].split(';')[0]
|
|
3116
|
-
encoding_scan = feed.meta['content-type'].to_s.scan(/charset\s*=\s*(.*?)(?:"|')*$/)
|
|
3117
|
-
http_encoding = encoding_scan.flatten[0].to_s.gsub(/("|')/,'')
|
|
3118
|
-
http_encoding = nil if http_encoding.empty?
|
|
3119
|
-
# FIXME Open-Uri returns iso8859-1 if there is no charset header,
|
|
3120
|
-
# but that doesn't pass the tests. Open-Uri claims its following
|
|
3121
|
-
# the right RFC. Are they wrong or do we need to change the tests?
|
|
3122
|
-
rescue NoMethodError
|
|
3123
|
-
http_headers = {}
|
|
3124
|
-
http_content_type = nil
|
|
3125
|
-
http_encoding = nil
|
|
3126
|
-
end
|
|
3127
|
-
# Must sniff for non-ASCII-compatible character encodings before
|
|
3128
|
-
# searching for XML declaration. This heuristic is defined in
|
|
3129
|
-
# section F of the XML specification:
|
|
3130
|
-
# http://www.w3.org/TR/REC-xml/#sec-guessing-no-ext-info
|
|
3131
|
-
begin
|
|
3132
|
-
if xml_data[0..3] == "\x4c\x6f\xa7\x94"
|
|
3133
|
-
# EBCDIC
|
|
3134
|
-
xml_data = _ebcdic_to_ascii(xml_data)
|
|
3135
|
-
elsif xml_data[0..3] == "\x00\x3c\x00\x3f"
|
|
3136
|
-
# UTF-16BE
|
|
3137
|
-
sniffed_xml_encoding = 'utf-16be'
|
|
3138
|
-
xml_data = uconvert(xml_data, 'utf-16be', 'utf-8')
|
|
3139
|
-
elsif xml_data.size >= 4 and xml_data[0..1] == "\xfe\xff" and xml_data[2..3] != "\x00\x00"
|
|
3140
|
-
# UTF-16BE with BOM
|
|
3141
|
-
sniffed_xml_encoding = 'utf-16be'
|
|
3142
|
-
xml_data = uconvert(xml_data[2..-1], 'utf-16be', 'utf-8')
|
|
3143
|
-
elsif xml_data[0..3] == "\x3c\x00\x3f\x00"
|
|
3144
|
-
# UTF-16LE
|
|
3145
|
-
sniffed_xml_encoding = 'utf-16le'
|
|
3146
|
-
xml_data = uconvert(xml_data, 'utf-16le', 'utf-8')
|
|
3147
|
-
elsif xml_data.size >=4 and xml_data[0..1] == "\xff\xfe" and xml_data[2..3] != "\x00\x00"
|
|
3148
|
-
# UTF-16LE with BOM
|
|
3149
|
-
sniffed_xml_encoding = 'utf-16le'
|
|
3150
|
-
xml_data = uconvert(xml_data[2..-1], 'utf-16le', 'utf-8')
|
|
3151
|
-
elsif xml_data[0..3] == "\x00\x00\x00\x3c"
|
|
3152
|
-
# UTF-32BE
|
|
3153
|
-
sniffed_xml_encoding = 'utf-32be'
|
|
3154
|
-
xml_data = uconvert(xml_data, 'utf-32be', 'utf-8')
|
|
3155
|
-
elsif xml_data[0..3] == "\x3c\x00\x00\x00"
|
|
3156
|
-
# UTF-32LE
|
|
3157
|
-
sniffed_xml_encoding = 'utf-32le'
|
|
3158
|
-
xml_data = uconvert(xml_data, 'utf-32le', 'utf-8')
|
|
3159
|
-
elsif xml_data[0..3] == "\x00\x00\xfe\xff"
|
|
3160
|
-
# UTF-32BE with BOM
|
|
3161
|
-
sniffed_xml_encoding = 'utf-32be'
|
|
3162
|
-
xml_data = uconvert(xml_data[4..-1], 'utf-32BE', 'utf-8')
|
|
3163
|
-
elsif xml_data[0..3] == "\xff\xfe\x00\x00"
|
|
3164
|
-
# UTF-32LE with BOM
|
|
3165
|
-
sniffed_xml_encoding = 'utf-32le'
|
|
3166
|
-
xml_data = uconvert(xml_data[4..-1], 'utf-32le', 'utf-8')
|
|
3167
|
-
elsif xml_data[0..2] == "\xef\xbb\xbf"
|
|
3168
|
-
# UTF-8 with BOM
|
|
3169
|
-
sniffed_xml_encoding = 'utf-8'
|
|
3170
|
-
xml_data = xml_data[3..-1]
|
|
3171
|
-
else
|
|
3172
|
-
# ASCII-compatible
|
|
3173
|
-
end
|
|
3174
|
-
xml_encoding_match = /^<\?.*encoding=[\'"](.*?)[\'"].*\?>/.match(xml_data)
|
|
3175
|
-
rescue
|
|
3176
|
-
xml_encoding_match = nil
|
|
3177
|
-
end
|
|
3178
|
-
if xml_encoding_match
|
|
3179
|
-
xml_encoding = xml_encoding_match[1].downcase
|
|
3180
|
-
xencodings = ['iso-10646-ucs-2', 'ucs-2', 'csunicode', 'iso-10646-ucs-4', 'ucs-4', 'csucs4', 'utf-16', 'utf-32', 'utf_16', 'utf_32', 'utf16', 'u16']
|
|
3181
|
-
if sniffed_xml_encoding and xencodings.include?xml_encoding
|
|
3182
|
-
xml_encoding = sniffed_xml_encoding
|
|
3183
|
-
end
|
|
3184
|
-
end
|
|
3185
|
-
|
|
3186
|
-
acceptable_content_type = false
|
|
3187
|
-
application_content_types = ['application/xml', 'application/xml-dtd', 'application/xml-external-parsed-entity']
|
|
3188
|
-
text_content_types = ['text/xml', 'text/xml-external-parsed-entity']
|
|
3189
|
-
|
|
3190
|
-
if application_content_types.include?(http_content_type) or (/^application\// =~ http_content_type and /\+xml$/ =~ http_content_type)
|
|
3191
|
-
acceptable_content_type = true
|
|
3192
|
-
true_encoding = http_encoding || xml_encoding || 'utf-8'
|
|
3193
|
-
elsif text_content_types.include?(http_content_type) or (/^text\// =~ http_content_type and /\+xml$/ =~ http_content_type)
|
|
3194
|
-
acceptable_content_type = true
|
|
3195
|
-
true_encoding = http_encoding || 'us-ascii'
|
|
3196
|
-
elsif /^text\// =~ http_content_type
|
|
3197
|
-
true_encoding = http_encoding || 'us-ascii'
|
|
3198
|
-
elsif http_headers and not http_headers.empty? and not http_headers.has_key?'content-type'
|
|
3199
|
-
true_encoding = xml_encoding || 'iso-8859-1'
|
|
3200
|
-
else
|
|
3201
|
-
true_encoding = xml_encoding || 'utf-8'
|
|
3202
|
-
end
|
|
3203
|
-
return true_encoding, http_encoding, xml_encoding, sniffed_xml_encoding, acceptable_content_type
|
|
3204
|
-
end
|
|
3205
|
-
|
|
3206
|
-
def self.toUTF8(data, encoding)
|
|
3207
|
-
=begin
|
|
3208
|
-
Changes an XML data stream on the fly to specify a new encoding
|
|
3209
|
-
|
|
3210
|
-
data is a raw sequence of bytes (not Unicode) that is presumed to be in %encoding already
|
|
3211
|
-
encoding is a string recognized by encodings.aliases
|
|
3212
|
-
=end
|
|
3213
|
-
$stderr << "entering self.toUTF8, trying encoding %s\n" % encoding if $debug
|
|
3214
|
-
# NOTE we must use double quotes when dealing with \x encodings!
|
|
3215
|
-
if (data.size >= 4 and data[0..1] == "\xfe\xff" and data[2..3] != "\x00\x00")
|
|
3216
|
-
if $debug
|
|
3217
|
-
$stderr << "stripping BOM\n"
|
|
3218
|
-
if encoding != 'utf-16be'
|
|
3219
|
-
$stderr << "string utf-16be instead\n"
|
|
3220
|
-
end
|
|
3221
|
-
end
|
|
3222
|
-
encoding = 'utf-16be'
|
|
3223
|
-
data = data[2..-1]
|
|
3224
|
-
elsif (data.size >= 4 and data[0..1] == "\xff\xfe" and data[2..3] != "\x00\x00")
|
|
3225
|
-
if $debug
|
|
3226
|
-
$stderr << "stripping BOM\n"
|
|
3227
|
-
$stderr << "trying utf-16le instead\n" if encoding != 'utf-16le'
|
|
3228
|
-
end
|
|
3229
|
-
encoding = 'utf-16le'
|
|
3230
|
-
data = data[2..-1]
|
|
3231
|
-
elsif (data[0..2] == "\xef\xbb\xbf")
|
|
3232
|
-
if $debug
|
|
3233
|
-
$stderr << "stripping BOM\n"
|
|
3234
|
-
$stderr << "trying utf-8 instead\n" if encoding != 'utf-8'
|
|
3235
|
-
end
|
|
3236
|
-
encoding = 'utf-8'
|
|
3237
|
-
data = data[3..-1]
|
|
3238
|
-
elsif (data[0..3] == "\x00\x00\xfe\xff")
|
|
3239
|
-
if $debug
|
|
3240
|
-
$stderr << "stripping BOM\n"
|
|
3241
|
-
if encoding != 'utf-32be'
|
|
3242
|
-
$stderr << "trying utf-32be instead\n"
|
|
3243
|
-
end
|
|
3244
|
-
end
|
|
3245
|
-
encoding = 'utf-32be'
|
|
3246
|
-
data = data[4..-1]
|
|
3247
|
-
elsif (data[0..3] == "\xff\xfe\x00\x00")
|
|
3248
|
-
if $debug
|
|
3249
|
-
$stderr << "stripping BOM\n"
|
|
3250
|
-
if encoding != 'utf-32le'
|
|
3251
|
-
$stderr << "trying utf-32le instead\n"
|
|
3252
|
-
end
|
|
3253
|
-
end
|
|
3254
|
-
encoding = 'utf-32le'
|
|
3255
|
-
data = data[4..-1]
|
|
3256
|
-
end
|
|
3257
|
-
begin
|
|
3258
|
-
newdata = uconvert(data, encoding, 'utf-8')
|
|
3259
|
-
rescue => details
|
|
3260
|
-
end
|
|
3261
|
-
$stderr << "successfully converted #{encoding} data to utf-8\n" if $debug
|
|
3262
|
-
declmatch = /^<\?xml[^>]*?>/
|
|
3263
|
-
newdecl = "<?xml version=\'1.0\' encoding=\'utf-8\'?>"
|
|
3264
|
-
if declmatch =~ newdata
|
|
3265
|
-
newdata.sub!(declmatch, newdecl)
|
|
3266
|
-
else
|
|
3267
|
-
newdata = newdecl + "\n" + newdata
|
|
3268
|
-
end
|
|
3269
|
-
return newdata
|
|
3270
|
-
end
|
|
3271
|
-
|
|
3272
|
-
def self.stripDoctype(data)
|
|
3273
|
-
=begin
|
|
3274
|
-
Strips DOCTYPE from XML document, returns (rss_version, stripped_data)
|
|
3275
|
-
|
|
3276
|
-
rss_version may be 'rss091n' or None
|
|
3277
|
-
stripped_data is the same XML document, minus the DOCTYPE
|
|
3278
|
-
=end
|
|
3279
|
-
entity_pattern = /<!ENTITY(.*?)>/m # m is for Regexp::MULTILINE
|
|
3280
|
-
data = data.gsub(entity_pattern,'')
|
|
3281
|
-
|
|
3282
|
-
doctype_pattern = /<!DOCTYPE(.*?)>/m
|
|
3283
|
-
doctype_results = data.scan(doctype_pattern)
|
|
3284
|
-
if doctype_results and doctype_results[0]
|
|
3285
|
-
doctype = doctype_results[0][0]
|
|
3286
|
-
else
|
|
3287
|
-
doctype = ''
|
|
3288
|
-
end
|
|
3289
|
-
|
|
3290
|
-
if /netscape/ =~ doctype.downcase
|
|
3291
|
-
version = 'rss091n'
|
|
3292
|
-
else
|
|
3293
|
-
version = nil
|
|
3294
|
-
end
|
|
3295
|
-
data = data.sub(doctype_pattern, '')
|
|
3296
|
-
return version, data
|
|
3297
|
-
end
|
|
3298
|
-
|
|
3299
|
-
def parse(*args); FeedParser.parse(*args); end
|
|
3300
|
-
def FeedParser.parse(furi, options={})
|
|
3301
|
-
# Parse a feed from a URL, file, stream or string
|
|
3302
|
-
$compatible = options[:compatible] || $compatible # Use the default compatibility if compatible is nil
|
|
3303
|
-
result = FeedParserDict.new
|
|
3304
|
-
result['feed'] = FeedParserDict.new
|
|
3305
|
-
result['entries'] = []
|
|
3306
|
-
if options[:modified]
|
|
3307
|
-
options[:modified] = Time.parse(options[:modified]).rfc2822
|
|
3308
|
-
# FIXME this ignores all of our time parsing work. Does it matter?
|
|
143
|
+
|
|
144
|
+
def parse(furi, options = {})
|
|
145
|
+
# Parse a feed from a URL, file, stream or string
|
|
146
|
+
$compatible = options[:compatible] || $compatible # Use the default compatibility if compatible is nil
|
|
147
|
+
strictklass = options[:strict] || StrictFeedParser
|
|
148
|
+
looseklass = options[:loose] || LooseFeedParser
|
|
149
|
+
result = FeedParserDict.new
|
|
150
|
+
result['feed'] = FeedParserDict.new
|
|
151
|
+
result['entries'] = []
|
|
152
|
+
if options[:modified]
|
|
153
|
+
options[:modified] = Time.parse(options[:modified]).rfc2822
|
|
154
|
+
# FIXME this ignores all of our time parsing work. Does it matter?
|
|
3309
155
|
end
|
|
3310
156
|
result['bozo'] = false
|
|
3311
157
|
handlers = options[:handlers]
|
|
3312
|
-
|
|
3313
158
|
if handlers.class != Array # FIXME why does this happen?
|
|
3314
159
|
handlers = [handlers]
|
|
3315
160
|
end
|
|
3316
161
|
|
|
3317
162
|
begin
|
|
3318
|
-
if
|
|
163
|
+
if File.exists?furi
|
|
3319
164
|
f = open(furi) # OpenURI doesn't behave well when passing HTTP options to a file.
|
|
3320
165
|
else
|
|
3321
166
|
# And when you do pass them, make sure they aren't just nil (this still true?)
|
|
@@ -3482,7 +327,7 @@ Strips DOCTYPE from XML document, returns (rss_version, stripped_data)
|
|
|
3482
327
|
if use_strict_parser
|
|
3483
328
|
# initialize the SAX parser
|
|
3484
329
|
saxparser = XML::SAX::Helpers::ParserFactory.makeParser("XML::Parser::SAXDriver")
|
|
3485
|
-
feedparser =
|
|
330
|
+
feedparser = strictklass.new(baseuri, baselang, 'utf-8')
|
|
3486
331
|
saxparser.setDocumentHandler(feedparser)
|
|
3487
332
|
saxparser.setDTDHandler(feedparser)
|
|
3488
333
|
saxparser.setEntityResolver(feedparser)
|
|
@@ -3503,7 +348,7 @@ Strips DOCTYPE from XML document, returns (rss_version, stripped_data)
|
|
|
3503
348
|
end
|
|
3504
349
|
end
|
|
3505
350
|
if not use_strict_parser
|
|
3506
|
-
feedparser =
|
|
351
|
+
feedparser = looseklass.new(baseuri, baselang, (known_encoding and 'utf-8' or ''))
|
|
3507
352
|
feedparser.parse(data)
|
|
3508
353
|
$stderr << "Using LooseFeed\n\n" if $debug
|
|
3509
354
|
end
|
|
@@ -3513,6 +358,7 @@ Strips DOCTYPE from XML document, returns (rss_version, stripped_data)
|
|
|
3513
358
|
result['namespaces'] = feedparser.namespacesInUse
|
|
3514
359
|
return result
|
|
3515
360
|
end
|
|
361
|
+
module_function(:parse)
|
|
3516
362
|
end # End FeedParser module
|
|
3517
363
|
|
|
3518
364
|
class Serializer
|
|
@@ -3552,7 +398,7 @@ class TextSerializer < Serializer
|
|
|
3552
398
|
end
|
|
3553
399
|
end
|
|
3554
400
|
|
|
3555
|
-
class PprintSerializer < Serializer # FIXME
|
|
401
|
+
class PprintSerializer < Serializer # FIXME use pp instead
|
|
3556
402
|
def write(stream = $stdout)
|
|
3557
403
|
stream << @results['href'].to_s + "\n\n"
|
|
3558
404
|
pp(@results)
|